[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M72T7P6CN7ST2TSEMTTH57PZ":3,"skill-same:01M72T7P6CN7ST2TSEMTTH57PZ":85},{"skillNo":4,"name":5,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":8,"repositoryNo":11,"starsCount":12,"licenseSpdxId":13,"licenseFileUrl":6,"sourceHtmlUrl":14,"categories":15,"keywords":16,"publishedAt":17,"linkOnly":18,"markdownContent":19,"markdownBody":20,"frontMatter":21,"refName":22,"commitSha":23,"skillDirectory":24,"forksCount":25,"repositoryHtmlUrl":26,"repositoryHomepageUrl":6,"ownerHtmlUrl":27,"ownerBlogUrl":28,"ownerDescription":6,"updatedAt":29,"install":30,"resources":43},"01M72T7P6CN7ST2TSEMTTH57PZ","ywc-spec-validate",null,"yongwoon\u002Fywc-agent-toolkit","yongwoon","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F9317029?v=4","01M72SZSF6VKANKG5QTMV4V95S","01M72SZSG8HHC3W2CV5P7EJ5AH",8,"MIT","https:\u002F\u002Fgithub.com\u002Fyongwoon\u002Fywc-agent-toolkit\u002Fblob\u002Fmain\u002Fclaude-code\u002Fskills\u002Fywc-spec-validate\u002FSKILL.md",[],[],"2026-11-10 22:39:42",false,"---\nname: ywc-spec-validate\nversion: 1.0.0\ndescription: (ywc) Use when a specification has been written and before task decomposition, and the user wants to validate spec completeness, consistency, feasibility, or readiness for implementation. Triggers: \"사양 검토\", \"spec review\", \"사양 리뷰\", \"review specification\", \"스펙 점검\", \"check the spec\", \"is this spec complete\", \"仕様レビュー\". Do not use for code-level review (use ywc-impl-review), product\u002Fbusiness review (use ywc-product-review), writing a new spec (use ywc-spec-writer), an automated validate-and-replan loop (use ywc-spec-ready), or when no specification document exists yet.\ncategory: spec\nphase: planning\nrequires: []\nadvisor_budget: 2\n---\n\n# ywc-spec-validate\n\n**Announce at start:** \"I'm using the ywc-spec-validate skill to validate specification completeness, consistency, and feasibility.\"\n\nSpec Reviewer Agent Skill for specification quality validation.\n\n## Rationalization Defense\n\nWhen tempted to skip a step, check this table first:\n\n| Excuse | Reality |\n|---|---|\n| \"Spec is mostly clear, no need to assess Feasibility\" | Feasibility and Code Compatibility require project context. Always read CLAUDE.md, package.json, and existing code first. |\n| \"Spec gaps are obvious, just list them and finish\" | Each gap must specify which dimension (completeness \u002F consistency \u002F feasibility \u002F compatibility) and propose a concrete clarifying question. |\n| \"User wrote the spec, do not push back too hard\" | Honest review prevents implementation rework. Soften tone, never soften findings. |\n| \"Spec uses internal jargon, infer the intent\" | If a term is undefined, that is a completeness gap. Flag it. |\n| \"Architecture decision is implicit, do not flag it\" | Implicit ≠ specified. Surface implicit assumptions for explicit confirmation. |\n| \"Spec follows best practices, so complexity is fine\" | The 4 base dimensions catch what is *missing* or *conflicting*, never what is *excessive*. Abstraction, configurability, or generality the stated scope does not yet require is over-engineering — surface it as a Simplicity Warning. Would a senior engineer call this spec overbuilt? |\n| \"The spec contradicts CLAUDE.md, follow the spec\" | Surface the conflict. Do not silently let the spec override project rules. |\n| \"4-dimension review is fast enough sequentially\" | Phase 1 fan-out cuts latency on large specs; each Sonnet subagent handles one dimension, reducing per-call context and preventing cross-dimension finding contamination. |\n| \"Spec passed and tasks were generated from it, so they must agree\" | task-generator can silently drop or reinterpret a requirement. When `--tasks` is set, build the Requirement Coverage table (Step 4c) — an UNCOVERED FR ships an incomplete feature while every spec-internal claim still validates cleanly. Drift between artifacts is invisible to spec-only review. |\n| \"Every spec deserves the full 2 advisor calls\" | Right-size the *advisor\u002Fcouncil* spend (not the cheap Sonnet fan-out): a single-section follow-up does not need the same Opus budget as a 15-AC multi-module spec. Spending advisor calls on a trivial spec wastes tokens; under-spending on a cross-cutting one misses real trade-offs. Apply the Complexity-Aware Budget Defaults before Phase 1. |\n| \"spec has no code yet, so concurrency is an implementation-phase problem\" | Missing concurrency\u002Ftransaction\u002Fidempotency requirements for a duplicate-sensitive write are a Completeness Critical — caught at the spec stage. |\n| \"The spec already describes the screen, no need to run the UI lens check\" | A screen description states what the UI looks like, not whether round-trip fidelity, loading\u002Fempty\u002Ferror states, or destructive-action reversibility are defined — Step 3.6 exists precisely because these are the QA expectations an author assumes are \"obvious\" and therefore never writes down. |\n\n**Violating the letter of these rules is violating the spirit.** A spec review that finds nothing is not a review.\n\n## Arguments\n\n| Parameter | Format | Example | Description |\n|-----------|--------|---------|-------------|\n| `--spec` | `--spec \u003Cpath>` | `--spec docs\u002Foutline\u002F02-api.md` | Specification file path (required) |\n| `--mode` | `--mode standard\\|socratic` | `--mode socratic` | Output style. `standard` (default) returns the finding report and gate verdict; `socratic` returns learning questions instead. See [references\u002Fsocratic-mode.md](references\u002Fsocratic-mode.md). |\n| `--focus` | `--focus \u003Carea>` | `--focus architecture` | Optional focus hint. When Council Escalation triggers, the generic 4 voices are replaced with domain expert profiles for the chosen area. Valid: `requirements`, `architecture`, `testing`, `compliance`. See [references\u002Fexpert-profiles.md](references\u002Fexpert-profiles.md). |\n| `--format` | `--format markdown\\|html` | `--format html` | Output format. Default `markdown`. With `html`, writes a self-contained HTML report to `claudedocs\u002F`. See [html-output.md](..\u002Freferences\u002Fhtml-output.md). Orthogonal to `--mode` |\n| `--advisor-budget` | `--advisor-budget \u003Cn>` | `--advisor-budget 0` | Opus advisor escalation budget for this invocation. Default: current behavior (≤2). `0` disables Phase 2 escalation entirely — ambiguous advisor candidates are reported as normal Suggestions (with a \"would have escalated but budget=0\" note) instead of being escalated. Used by orchestrators (e.g. `ywc-spec-ready`) to cap cumulative advisor cost across iterations. |\n| `--tasks` | `--tasks \u003Cdir>` | `--tasks tasks\u002F` | Optional. When set, runs the **Cross-Artifact Consistency (Analyze)** pass (Step 4c) after Phase 1: verifies the spec and the generated task directory have not drifted apart — every requirement covered by a task, no orphan task, dependency order respects spec phases. Omit for the default pre-decomposition review (tasks do not exist yet). |\n\n## Execution Steps\n\n1. **Collect Project Context** — Read the following before touching the spec file. The Feasibility and Code Compatibility dimensions cannot be assessed without this context:\n   - `CLAUDE.md` (or the nearest parent `CLAUDE.md`) — language\u002Fframework constraints, commit conventions, any explicit out-of-scope rules\n   - `package.json` \u002F `pyproject.toml` \u002F `go.mod` — runtime, major dependencies, and their versions\n   - Top-level directory structure — identify where DB schemas, API routes, and type definitions live (e.g., `src\u002F`, `app\u002F`, `prisma\u002F`, `migrations\u002F`)\n   - `docs\u002Fubiquitous-language.md` (if it exists) — canonical domain terms and their \"Synonyms to Avoid\"; any spec term matching a Synonyms entry is a Consistency finding\n   - Extract and keep: runtime, primary framework, DB type, ORM, existing API pattern (REST \u002F GraphQL \u002F RPC)\n\n   If CLAUDE.md is absent and package.json is absent, note \"project context unavailable\" and flag all Feasibility findings as lower-confidence.\n\n2. **Read Specification File** — Confirm the `--spec` path exists before reading. If the file does not exist, stop immediately with `BLOCKED` status and report the missing path.\n\n3. **Check Code Compatibility** — Use Grep\u002FGlob to search for related DB schemas, API routes, and type definitions to detect conflicts with the spec. Scope the search to the directories identified in Step 1.\n\n3.5. **Precedent-Completeness Re-grounding (MANDATORY omission check — distinct from the conflict check in Step 3).** Step 3 \u002F Code Compatibility detect where a spec *claim* conflicts with code. They structurally cannot detect an *omission*: an existing integration site the spec never names makes no claim, so nothing conflicts — yet a dropped integration site silently breaks the feature on the un-modeled runtime path while every written claim still validates cleanly. (This is the failure class that survives repeated spec-validation; re-reading spec text can never reveal a site the text never names.)\n\n   For **every** existing pattern the spec mirrors \u002F follows \u002F 踏襲 \u002F extends (a **named precedent X**), you MUST build a **Precedent Site Coverage table** before writing the report — do not reason about it in prose, build the table:\n\n   1. Grep X's own identifier(s)\u002Fmarker(s) across the actual code: `grep -rn \"\u003CprecedentFn>\\|\u003CprecedentMarker>\" \u003Csrc>` (exclude test files). Integration patterns (snippet\u002Fscript injection, hooks, interceptors, lifecycle handlers) routinely span a **generation** step, a **preview\u002Ftransform** step, **and** a **publish\u002Frender-time** step — enumerate all of them; the publish\u002Frender-time site is the one a memory-driven spec most often drops.\n   2. Emit one row **per site X integrates at**:\n\n      | Precedent site (`file:line` + what runs there) | Spec coverage: Replicated \u002F Deferred-with-reason \u002F **OMITTED** |\n\n   3. Every grep hit MUST appear as a row. Decide each row's coverage by searching the spec for *that specific site* — never by trusting the spec's own prose summary that it \"follows X\".\n   4. **Every row marked OMITTED is its own Completeness Critical.** Do not fold multiple omitted sites into one finding, and **do not let a louder conflict elsewhere (e.g. a stale-anchor or deprecated-path finding) substitute for explicitly naming the omitted site** — an omission and a conflict are different defects and both must be reported. Each OMITTED Critical names the omitted site's `file:line` and the runtime path on which the feature fails.\n\n   Include the coverage table in the report, and feed each OMITTED row into the Completeness dimension of Phase 1 as a separate Critical.\n\n3.6. **UI\u002FInteraction Omission Check (MANDATORY omission check).** Step 3.5 targets precedent-site omissions; this step targets UI\u002Finteraction trigger surfaces with the same structure — a spec's own prose rarely states a QA expectation an experienced reviewer would apply by habit, so the check runs against a fixed lens table rather than the spec's own claims.\n\n   > **Action required**: Read [references\u002Fui-qa-lenses.md](references\u002Fui-qa-lenses.md) now for the canonical lens table before running this check.\n\n   For each lens row whose Trigger surface matches this spec, check whether the spec's Acceptance Criteria \u002F Functional Requirements \u002F Edge Cases satisfy that lens's Rule. A triggered lens the spec does not address is a **Completeness Warning** — cite the lens name and the nearest relevant spec section (or its absence). Feed each such row into the Completeness dimension of Phase 1 as its own finding (do not fold several lenses into one), and include them in the report. Skip any lens whose trigger surface the spec does not touch; do not force-fit lenses onto a non-UI spec.\n\n4. **Phase 1 — Parallel Dimension Review** — Use the Task tool to spawn 4 Sonnet subagents in parallel, one per review dimension. Pass each subagent the project context from Step 1 and the spec text from Step 2:\n\n   > **Action required**: Read [..\u002Freferences\u002Fsubagent-async-monitoring.md](..\u002Freferences\u002Fsubagent-async-monitoring.md) now — these four dimension subagents are one batch under that contract (distinct `name` per dimension, full-roster reconciliation on every signal, turn-boundary waiting, bounded escalation). Apply it before aggregating the dimension returns. **Release (mechanical gate — do not skip)**: named dispatches stay `idle` after finishing instead of exiting. After all four dimension returns are aggregated, call `TaskStop` once per dimension subagent with its `name` as `task_id` (current batch only; skip any still `running`). See the \"Releasing Completed Named Agents\" section of the reference.\n\n   | Subagent | Model | Dimension | Focus |\n   |---|---|---|---|\n   | Completeness | sonnet | Completeness | Missing required items, edge cases, pagination. For a **duplicate-sensitive write flow** (stock\u002Fbalance\u002Forder\u002Fpayment\u002Fprovisioning\u002Fquota) also check whether the spec specifies: (a) concurrent-request behavior, (b) the transaction\u002Fconsistency boundary spanning multiple DB writes, (c) handling of duplicate client retry \u002F double-click, (d) the failure response\u002Fstatus when stock\u002Fbalance\u002Fquota is exhausted or a lock\u002Fversion mismatch occurs. Missing any → a Completeness finding: **Critical** when the gap can cause double charge \u002F oversell \u002F lost ledger \u002F duplicate provisioning, otherwise **Warning** |\n   | Consistency | sonnet | Consistency | Terminology mismatches; Synonyms-to-Avoid violations |\n   | Feasibility | sonnet | Feasibility + Simplicity | Implementable with current stack and dependencies; also flag over-engineering — abstraction, configurability, or generality the stated scope does not yet require (surface as Warning). Check command-bearing AC with the AC baseline check (see the callout below the table). |\n   | Code Compatibility | sonnet | Code Compatibility | Conflicts with existing schema, API routes, type definitions |\n\n   > **Action required**: Read [references\u002Fac-baseline-check.md](references\u002Fac-baseline-check.md) now and pass its content to the Feasibility subagent — it defines the path exclusions, the Glob-only existence check, the single Grep-tool-call re-run and the `Baseline:` comparison.\n\n   Each subagent returns:\n   - **Confirmed findings** — dimension label, severity (Critical \u002F Warning \u002F Suggestion), file:line, description, improvement\n   - **Advisor candidates** — findings where two reasonable interpretations exist (include the specific choice and its consequence, ≤100 lines per candidate)\n\n   **Step 4b — Aggregate**: Combine and deduplicate findings by `{file}:{line}`. Cap advisor candidates at `advisor_budget` (default: 2), prioritizing Critical over Warning. Log any dropped candidates in the report.\n\n   The 4-dimension analysis runs the same way regardless of `--mode`; the mode only changes how findings are presented in step 5.\n\n4c. **Cross-Artifact Consistency (Analyze) — only when `--tasks \u003Cdir>` is set.** Steps 1–4 validate the spec in isolation. When a task directory already exists (re-validation after `ywc-task-generator`, or inside a convergence loop), verify the spec and tasks have not drifted apart — a defect class that spec-internal review structurally cannot see (an uncovered requirement makes no claim, so nothing conflicts, yet the feature ships incomplete). Build two tables before writing the report:\n\n   **Requirement Coverage** — one row per spec Functional Requirement \u002F Acceptance Criterion. Grep the task directory for each FR\u002FAC:\n\n   | Spec FR\u002FAC (`file:line`) | Covering task(s) (`tasks\u002F\u003Cid>`) | Replicated \u002F Deferred-with-reason \u002F **UNCOVERED** |\n\n   Every **UNCOVERED** row is its own Completeness Critical (a requirement no task implements).\n\n   **Task Provenance** — one row per task. Trace each task back to the spec:\n\n   | Task (`tasks\u002F\u003Cid>`) | Spec basis (`file:line` FR\u002FAC or named precedent) \u002F **ORPHAN** |\n\n   Every **ORPHAN** row (a task with no spec basis) is a Consistency Warning (unspecified scope). Then confirm the task dependency graph respects the spec's declared phase\u002Fsequence ordering — a task depending on a later-phase task is a Consistency finding.\n\n   Feed UNCOVERED rows into the Completeness dimension and ORPHAN \u002F dependency-order rows into the Consistency dimension as separate findings (do not fold multiple into one). Include both tables in the report.\n\n5. **Output Report — Branch on `--mode`**\n   - `standard` (default): emit the severity-classified finding report in the format below.\n   - `socratic`: emit a learning-question list per [references\u002Fsocratic-mode.md](references\u002Fsocratic-mode.md). Findings still drive question selection — they are just reshaped from \"fix X\" into \"what would happen if X?\". Completion Status is `SOCRATIC` and the gate score is informational only.\n\n## Review Dimensions\n\n| Dimension | Review Focus |\n|-----------|-------------|\n| Completeness | Missing required items (error handling, edge cases, pagination, etc.) |\n| Consistency | Terminology, format, or data structure mismatches across documents; spec terms appearing in the \"Synonyms to Avoid\" column of `docs\u002Fubiquitous-language.md` |\n| Feasibility | Whether the spec can be implemented with the current stack, and whether its command-bearing AC pass the [AC baseline check](references\u002Fac-baseline-check.md) |\n| Code Compatibility | Conflicts with existing DB schema, API route patterns |\n| Simplicity | Whether the spec specifies abstraction, configurability, or generality the stated scope does not yet require (over-engineering). Assessed within the Feasibility pass — surface speculative scope as a Warning, not a separate subagent. |\n\nUI\u002FInteraction omission mapping (Step 3.6):\n- **Completeness**: the spec triggers ≥1 lens from [references\u002Fui-qa-lenses.md](references\u002Fui-qa-lenses.md) but its Acceptance Criteria \u002F Functional Requirements \u002F Edge Cases do not satisfy that lens's rule.\n\n## Output Format\n\n```text\n## Spec Review Result: {filename}\n\n### Summary\n- Critical: N, Warning: M, Suggestion: K\n- Phase 2 advisor calls used: X of N ({single-advisor|council|none})   (N = applied budget; default 2, or the `--advisor-budget` value)\n\n### Precedent Site Coverage\n| Precedent site (`file:line` + what runs there) | Spec coverage: Replicated \u002F Deferred-with-reason \u002F OMITTED |\n|---|---|\n| ... | ... |\n\n### Critical Issues\n1. [{file}:{line}] [P1|P2] Description — Suggested improvement\n   (if P2) Advisor rationale: {one-line rationale}\n\n### Warning Issues\n1. [{file}:{line}] [P1|P2] Description — Suggested improvement\n\n### Suggestions\n1. [{file}:{line}] Description — Suggested improvement\n\n### Completion Status\n(One of: DONE | DONE_WITH_CONCERNS | BLOCKED | NEEDS_CONTEXT)\n```\n\n**Completion Status rules:**\n\nStatuses decided by the Confidence Gate result (`DONE`, `DONE_WITH_CONCERNS`, and the gate's `NEEDS_CONTEXT` \u002F `BLOCKED` bands) are defined only in the Band-to-status mapping table under [Confidence Gate](#confidence-gate) — that table is the single source; do not restate it here. The table below lists the pre-gate statuses, which are decided before (or instead of) the gate runs:\n\n| Pre-gate status | When to use |\n|--------|------------|\n| `BLOCKED` | Review cannot proceed — spec file missing, or a Phase 2 advisor returned an inconclusive verdict |\n| `NEEDS_CONTEXT` | `--spec` argument is missing or the file is empty\u002Funreadable |\n| `SOCRATIC` | `--mode socratic` was used; output is a learning-question list, not a gate verdict. Downstream skills (especially `ywc-task-generator`) must not consume this status as a handoff signal. |\n\n**Precedence — read this before emitting a status.** The Pre-gate status table above and the\nConfidence Gate's Band-to-status mapping are **not two independent rules**; the gate decision\nstatus comes only from the Band-to-status table. A spec with zero Critical findings scoring 78 is\n`NEEDS_CONTEXT` by the band, never `DONE`. Resolve it this way:\n\n1. **The Confidence Gate band decides the status.** It is the outer gate; run it last and let it win.\n2. **The finding count only splits the PROCEED band** — `DONE` when no Critical findings remain,\n   `DONE_WITH_CONCERNS` when any do. Below PROCEED the finding count does not upgrade the status:\n   a zero-Critical spec in REVIEW is `NEEDS_CONTEXT`, and in STOP it is `BLOCKED`.\n3. **Input failures short-circuit both.** A missing\u002Fempty\u002Funreadable `--spec` is `NEEDS_CONTEXT`, and\n   a missing spec file or an inconclusive Phase 2 advisor is `BLOCKED` — emitted *without* computing\n   a gate score, because there is nothing to score.\n\nSo `NEEDS_CONTEXT` and `BLOCKED` each have two legitimate triggers: an input failure (rule 3) or a\ngate band (rule 1). That overlap is intentional, not a conflict — but always say which one fired, so\na caller can tell \"I could not read your spec\" from \"I read it and it is not ready\".\n\n`SOCRATIC` sits outside this hierarchy: it is a mode, not a verdict, and no gate band produces it.\n\n**Re-plan handoff on `DONE_WITH_CONCERNS`** — after printing the report body, append the following one-liner so the user (or an interactive orchestrator) can resolve findings via `ywc-plan` Re-plan Mode instead of rewriting the spec from scratch:\n\n```\nTo address the Critical findings above without losing validated sections, run:\n\u002Fywc-plan --update-spec \u003Cspec-path> --failure-context \"\u003Cone-paragraph summary of the Critical findings>\"\n```\n\nThis routes findings into the `## Iteration N Amendments` append flow described in `ywc-plan` Step 4c, preserving validated portions of the spec. Print this line only on `DONE_WITH_CONCERNS` — `DONE`, `BLOCKED`, `NEEDS_CONTEXT`, and `SOCRATIC` do not benefit from Re-plan Mode. For programmatic consumers (e.g., `ywc-agentic`), the Programmatic Consumer Policy below already invokes the equivalent flow; this surface is for interactive users who would otherwise re-write from scratch.\n\n> **HTML mode (`--format html`)** — emits the same findings as a self-contained HTML report: severity color coding, tab navigation, and a `Copy as Markdown` button. Structure and conventions follow [html-output.md](..\u002Freferences\u002Fhtml-output.md). The Markdown surface is preserved inside the file, so downstream integration is unaffected.\n\n## Advisor Escalation Policy\n\nThis skill runs Phase 1 as 4 parallel Sonnet subagents (one per dimension) and aggregates their findings. For a small number of **genuinely ambiguous** findings from Phase 1, the orchestrator escalates to an Opus advisor using the Task tool with `model: opus`. This follows **Pattern B** from [advisor-pattern.md](..\u002Freferences\u002Fadvisor-pattern.md) — parallel executors in Phase 1, frontier judgment applied only where it actually matters in Phase 2.\n\n**Budget**: up to 2 Opus advisor calls per invocation by default. Unused budget is good. Spec review is smaller in scope than impl-review, so the cap is tighter.\n\n**Complexity-Aware Budget Defaults** — right-size the advisor\u002Fcouncil spend to the spec's scope *before* Phase 1 (an explicit `--advisor-budget` always overrides this default):\n\n| Spec scope | Default advisor budget |\n|---|---|\n| Single-section follow-up \u002F ≤2 modules touched | 0–1 |\n| Standard spec \u002F 3–5 modules | 2 (current default) |\n| Cross-cutting \u002F ≥15 acceptance criteria or >5 modules | 2, and prefer Council Escalation for the high-stakes finding |\n\nThe Sonnet 4-dimension fan-out (Phase 1) always runs regardless of scope — it is cheap and catches omissions. Only the Opus Phase 2 \u002F Council spend is right-sized here, so the \"Right Job, Right Tool\" saving never comes at the cost of missed findings.\n\n**`--advisor-budget \u003Cn>` override**: when set, `\u003Cn>` is the hard ceiling on Phase 2 Opus calls for this invocation (default 2). The report header's `Phase 2 advisor calls used: X of N` line reflects the applied budget as `N` so an orchestrator can track cumulative spend across iterations. When `--advisor-budget 0`, Phase 2 escalation is **disabled entirely**: each finding that would have escalated is instead reported as a normal Suggestion carrying a one-line `would have escalated but budget=0` rationale — it is **never silently dropped**. This lets a loop driver (e.g. `ywc-spec-ready`) run cheap deterministic passes and spend advisor budget only when it chooses to.\n\n**Escalation conditions** — each must satisfy all three properties from [advisor-pattern.md §5](..\u002Freferences\u002Fadvisor-pattern.md) (objective trigger, irreversibility, ambiguity):\n\n1. **Feasibility uncertainty** — The spec appears feasible but hinges on an assumption about an undocumented library feature, version-specific behavior, or performance characteristic. A wrong assumption would make the spec unimplementable or force mid-implementation scope cuts.\n2. **Code Compatibility conflict with plausible-either-way resolution** — The spec conflicts with existing code (schema, API contract, or type definition), and both \"spec adapts to code\" and \"code refactors to spec\" have plausible precedent in the project. The advisor picks which side moves.\n3. **Completeness borderline** — The spec omits a feature that could be intentional (out of scope, deferred to another spec) or an oversight. The wrong verdict delays implementation (if flagged as gap when actually out of scope) or generates incorrect downstream tasks (if dismissed when actually a gap).\n4. **Cross-document consistency conflict** — The spec conflicts with another spec or architecture document in the project, and resolving the conflict requires domain judgment about which document is the current source of truth.\n\n**Context payload rules** (critical for cost discipline):\n- Forward only the decision: the specific spec clause, the conflicting code or spec excerpt, and the constraint (≤100 lines total).\n- Do NOT forward: the full spec, the full target code, or the CLAUDE.md contents.\n- The advisor returns a short verdict (≤200 words) containing the recommended resolution and a one-line rationale.\n\n**Non-goals** — do not escalate for these:\n- Terminology nitpicks or formatting inconsistencies (these are Phase 1 Warnings, not advisor material).\n- Generic \"the spec could be clearer\" observations without a specific failure mode.\n- Stylistic preferences about how requirements are expressed.\n- Missing sections that are clearly an author oversight (just report it as a Warning).\n\nReport escalations in the output: mark Phase 2 findings with `[P2]` prefix and include the advisor's one-line rationale. This preserves auditability of which findings involved frontier judgment.\n\n## Council Escalation (High-Stakes Trade-offs)\n\nWhen a finding involves a genuine architectural trade-off — where two reasonable positions exist and the wrong choice has significant implementation cost — the single Opus advisor may be replaced by a 4-voice **Council** review. Use council escalation when all three conditions hold:\n\n1. The trade-off has long-term architectural consequences (e.g., sync vs. async boundary, data ownership model, API versioning strategy).\n2. The single-advisor verdict would likely depend on unstated assumptions about team preferences or future requirements.\n3. The finding is Critical or Warning severity — not a Suggestion.\n\n**Council voices** (spawn one Opus subagent per voice, passing only the specific trade-off excerpt — not the full spec):\n\n| Voice | Role | Anti-bias Mandate |\n|-------|------|-------------------|\n| Architect | Correctness and long-term maintainability | Challenge convenience-driven choices |\n| Skeptic | Break the dominant assumption; find what everyone is taking for granted | Ignore sunk cost arguments |\n| Pragmatist | Shipping speed and team execution risk | Ignore theoretical purity |\n| Critic | Failure modes; what breaks at scale or under adversarial conditions | Assume the happy path is wrong |\n\n**Anti-anchoring rule**: Each voice receives only the trade-off description — not the other voices' verdicts. Collect all four independently before synthesizing.\n\n> **Action required**: Read [..\u002Freferences\u002Fsubagent-async-monitoring.md](..\u002Freferences\u002Fsubagent-async-monitoring.md) now — the four Council voices are one batch under that contract (distinct `name` per voice, full-roster reconciliation on every signal, turn-boundary waiting, bounded escalation). **Release (mechanical gate — do not skip)**: named dispatches stay `idle` after finishing instead of exiting. After all four voice verdicts are collected and synthesized, call `TaskStop` once per voice subagent with its `name` as `task_id` (current batch only; skip any still `running`). See the \"Releasing Completed Named Agents\" section of the reference.\n\n**Council output format** (append to the relevant finding in the report):\n```\n[COUNCIL] {trade-off description}\n- Architect: {verdict, ≤2 sentences}\n- Skeptic: {verdict, ≤2 sentences}\n- Pragmatist: {verdict, ≤2 sentences}\n- Critic: {verdict, ≤2 sentences}\nConsensus: {agreed recommendation, or \"No consensus — strongest dissent: \u003Cvoice>: \u003Creason>\"}\n```\n\n### Profile Override (Focus-Specific Voices)\n\nWhen `--focus \u003Carea>` is set on the invocation and Council Escalation triggers, replace the 4 generic voices above with 3–4 domain expert profiles from [references\u002Fexpert-profiles.md](references\u002Fexpert-profiles.md). The Anti-anchoring rule, output format, and budget rule all apply unchanged — only the voice labels and heuristics change.\n\n| `--focus` value | Profiles to use (in order) |\n|---|---|\n| `requirements` | Requirements Quality + Specification by Example + Use Case Modeling |\n| `architecture` | Interface Design + Service Boundaries + Integration Patterns + Production Resilience |\n| `testing` | Risk-Based Testing + Three Amigos + Quality Attribute |\n| `compliance` | Production Resilience + Cloud-Native Operations + Audit Trail |\n\nUse generic Council (no `--focus`) when the trade-off spans two or more areas, or when the spec covers multiple domains and no single focus dominates. Mixing generic voices and focus profiles within a single Council session breaks anti-anchoring — pick one mode per session.\n\nCouncil escalation consumes the full 2-advisor budget. Do not use council and single-advisor escalation in the same invocation.\n\n## Confidence Gate\n\nThis skill applies the [Confidence Gate](..\u002Freferences\u002Fconfidence-gate.md) before declaring the spec ready for `task-generator`.\n\n**Required dimensions** (must each score ≥ 70):\n\n- **Scope clarity** — A spec with undefined scope cannot be decomposed into tasks. If scope is fuzzy, the gate forces STOP.\n- **Root cause identified** — The user need behind the spec must be named. Surface-level requirements without a stated user need produce tasks that solve the wrong problem.\n\n**Band-to-status mapping** for this skill (conforms to [confidence-gate.md §5](..\u002Freferences\u002Fconfidence-gate.md) — PROCEED splits on Critical findings so `DONE_WITH_CONCERNS` stays reachable):\n\n| Gate band | Completion status | Action |\n|-----------|-------------------|--------|\n| PROCEED (≥ 90), no Critical findings | DONE | Spec is ready for `task-generator`. |\n| PROCEED (≥ 90), Critical findings present | DONE_WITH_CONCERNS | Resolve the Critical findings via the Re-plan handoff before `task-generator`. |\n| REVIEW (70 – 89) | NEEDS_CONTEXT | Present 1-3 open questions; consider council escalation if multiple dimensions are \u003C 80. |\n| STOP (\u003C 70) | BLOCKED | Report which dimensions are weak; do not hand off to `task-generator`. |\n\nThe gate score must appear in the report header alongside the Critical\u002FWarning\u002FSuggestion finding counts. A required dimension scoring below 50 forces STOP regardless of aggregate.\n\n## Integration\n\n- **Primary upstream** — `ywc-spec-writer`: validates `docs\u002Fspecification\u002F` section files (`01-overview.md` … `07-glossary.md`). All 4 review dimensions (completeness, consistency, feasibility, code compatibility) apply fully.\n- **Secondary upstream** — `ywc-plan`: can validate feature plan documents in `docs\u002Fywc-plans\u002F`. Completeness and consistency dimensions apply; feasibility and code compatibility apply partially because plan documents are less implementation-specific.\n- **Cross-artifact (Analyze)** — `ywc-task-generator` output (`tasks\u002F`): once tasks exist, pass the directory via `--tasks \u003Cdir>` (alongside `--spec`) to run the Cross-Artifact Consistency pass (Step 4c) — confirms every spec requirement is covered by a task and no task is orphaned. Passing a task directory as `--spec` is still a misuse; `--tasks` is the correct channel.\n- **Downstream**: `ywc-task-generator` — hand off only when completion status is `DONE`. `DONE_WITH_CONCERNS` requires spec revision first.\n\n### Programmatic Consumer Policy\n\nWhen consumed by an orchestrator that cannot prompt for human input (e.g., ywc-agentic):\n\n| Completion Status | Action |\n| --- | --- |\n| DONE | Proceed to ywc-task-generator |\n| DONE_WITH_CONCERNS | Re-invoke ywc-plan with --update-spec + --failure-context + --non-interactive (max 1 retry), then re-validate |\n| BLOCKED | Stop execution; surface a structured triage report containing: (1) attempted triage steps, (2) verbatim blocker text, (3) proposed default action |\n| NEEDS_CONTEXT | Stop execution; surface a structured triage report containing: (1) attempted triage steps, (2) verbatim blocker text, (3) proposed default action |\n| SOCRATIC | Stop execution and report to user — SOCRATIC output is not a handoff signal; re-run without `--mode socratic` to obtain a gate verdict |\n\n> **Looped consumer**: the table's `max 1 retry` is the default for non-loop consumers (e.g. `ywc-agentic` Step 3 historically). When **multi-iteration** convergence is required, the dedicated `ywc-spec-ready` skill drives the `DONE_WITH_CONCERNS → ywc-plan --update-spec → re-validate` cycle repeatedly under its own `--max-iterations` cap and stall guards, and passes the remaining advisor budget via `--advisor-budget`. Use `ywc-spec-ready` instead of hand-rolling more than one retry.\n","# ywc-spec-validate\n\n**Announce at start:** \"I'm using the ywc-spec-validate skill to validate specification completeness, consistency, and feasibility.\"\n\nSpec Reviewer Agent Skill for specification quality validation.\n\n## Rationalization Defense\n\nWhen tempted to skip a step, check this table first:\n\n| Excuse | Reality |\n|---|---|\n| \"Spec is mostly clear, no need to assess Feasibility\" | Feasibility and Code Compatibility require project context. Always read CLAUDE.md, package.json, and existing code first. |\n| \"Spec gaps are obvious, just list them and finish\" | Each gap must specify which dimension (completeness \u002F consistency \u002F feasibility \u002F compatibility) and propose a concrete clarifying question. |\n| \"User wrote the spec, do not push back too hard\" | Honest review prevents implementation rework. Soften tone, never soften findings. |\n| \"Spec uses internal jargon, infer the intent\" | If a term is undefined, that is a completeness gap. Flag it. |\n| \"Architecture decision is implicit, do not flag it\" | Implicit ≠ specified. Surface implicit assumptions for explicit confirmation. |\n| \"Spec follows best practices, so complexity is fine\" | The 4 base dimensions catch what is *missing* or *conflicting*, never what is *excessive*. Abstraction, configurability, or generality the stated scope does not yet require is over-engineering — surface it as a Simplicity Warning. Would a senior engineer call this spec overbuilt? |\n| \"The spec contradicts CLAUDE.md, follow the spec\" | Surface the conflict. Do not silently let the spec override project rules. |\n| \"4-dimension review is fast enough sequentially\" | Phase 1 fan-out cuts latency on large specs; each Sonnet subagent handles one dimension, reducing per-call context and preventing cross-dimension finding contamination. |\n| \"Spec passed and tasks were generated from it, so they must agree\" | task-generator can silently drop or reinterpret a requirement. When `--tasks` is set, build the Requirement Coverage table (Step 4c) — an UNCOVERED FR ships an incomplete feature while every spec-internal claim still validates cleanly. Drift between artifacts is invisible to spec-only review. |\n| \"Every spec deserves the full 2 advisor calls\" | Right-size the *advisor\u002Fcouncil* spend (not the cheap Sonnet fan-out): a single-section follow-up does not need the same Opus budget as a 15-AC multi-module spec. Spending advisor calls on a trivial spec wastes tokens; under-spending on a cross-cutting one misses real trade-offs. Apply the Complexity-Aware Budget Defaults before Phase 1. |\n| \"spec has no code yet, so concurrency is an implementation-phase problem\" | Missing concurrency\u002Ftransaction\u002Fidempotency requirements for a duplicate-sensitive write are a Completeness Critical — caught at the spec stage. |\n| \"The spec already describes the screen, no need to run the UI lens check\" | A screen description states what the UI looks like, not whether round-trip fidelity, loading\u002Fempty\u002Ferror states, or destructive-action reversibility are defined — Step 3.6 exists precisely because these are the QA expectations an author assumes are \"obvious\" and therefore never writes down. |\n\n**Violating the letter of these rules is violating the spirit.** A spec review that finds nothing is not a review.\n\n## Arguments\n\n| Parameter | Format | Example | Description |\n|-----------|--------|---------|-------------|\n| `--spec` | `--spec \u003Cpath>` | `--spec docs\u002Foutline\u002F02-api.md` | Specification file path (required) |\n| `--mode` | `--mode standard\\|socratic` | `--mode socratic` | Output style. `standard` (default) returns the finding report and gate verdict; `socratic` returns learning questions instead. See [references\u002Fsocratic-mode.md](references\u002Fsocratic-mode.md). |\n| `--focus` | `--focus \u003Carea>` | `--focus architecture` | Optional focus hint. When Council Escalation triggers, the generic 4 voices are replaced with domain expert profiles for the chosen area. Valid: `requirements`, `architecture`, `testing`, `compliance`. See [references\u002Fexpert-profiles.md](references\u002Fexpert-profiles.md). |\n| `--format` | `--format markdown\\|html` | `--format html` | Output format. Default `markdown`. With `html`, writes a self-contained HTML report to `claudedocs\u002F`. See [html-output.md](..\u002Freferences\u002Fhtml-output.md). Orthogonal to `--mode` |\n| `--advisor-budget` | `--advisor-budget \u003Cn>` | `--advisor-budget 0` | Opus advisor escalation budget for this invocation. Default: current behavior (≤2). `0` disables Phase 2 escalation entirely — ambiguous advisor candidates are reported as normal Suggestions (with a \"would have escalated but budget=0\" note) instead of being escalated. Used by orchestrators (e.g. `ywc-spec-ready`) to cap cumulative advisor cost across iterations. |\n| `--tasks` | `--tasks \u003Cdir>` | `--tasks tasks\u002F` | Optional. When set, runs the **Cross-Artifact Consistency (Analyze)** pass (Step 4c) after Phase 1: verifies the spec and the generated task directory have not drifted apart — every requirement covered by a task, no orphan task, dependency order respects spec phases. Omit for the default pre-decomposition review (tasks do not exist yet). |\n\n## Execution Steps\n\n1. **Collect Project Context** — Read the following before touching the spec file. The Feasibility and Code Compatibility dimensions cannot be assessed without this context:\n   - `CLAUDE.md` (or the nearest parent `CLAUDE.md`) — language\u002Fframework constraints, commit conventions, any explicit out-of-scope rules\n   - `package.json` \u002F `pyproject.toml` \u002F `go.mod` — runtime, major dependencies, and their versions\n   - Top-level directory structure — identify where DB schemas, API routes, and type definitions live (e.g., `src\u002F`, `app\u002F`, `prisma\u002F`, `migrations\u002F`)\n   - `docs\u002Fubiquitous-language.md` (if it exists) — canonical domain terms and their \"Synonyms to Avoid\"; any spec term matching a Synonyms entry is a Consistency finding\n   - Extract and keep: runtime, primary framework, DB type, ORM, existing API pattern (REST \u002F GraphQL \u002F RPC)\n\n   If CLAUDE.md is absent and package.json is absent, note \"project context unavailable\" and flag all Feasibility findings as lower-confidence.\n\n2. **Read Specification File** — Confirm the `--spec` path exists before reading. If the file does not exist, stop immediately with `BLOCKED` status and report the missing path.\n\n3. **Check Code Compatibility** — Use Grep\u002FGlob to search for related DB schemas, API routes, and type definitions to detect conflicts with the spec. Scope the search to the directories identified in Step 1.\n\n3.5. **Precedent-Completeness Re-grounding (MANDATORY omission check — distinct from the conflict check in Step 3).** Step 3 \u002F Code Compatibility detect where a spec *claim* conflicts with code. They structurally cannot detect an *omission*: an existing integration site the spec never names makes no claim, so nothing conflicts — yet a dropped integration site silently breaks the feature on the un-modeled runtime path while every written claim still validates cleanly. (This is the failure class that survives repeated spec-validation; re-reading spec text can never reveal a site the text never names.)\n\n   For **every** existing pattern the spec mirrors \u002F follows \u002F 踏襲 \u002F extends (a **named precedent X**), you MUST build a **Precedent Site Coverage table** before writing the report — do not reason about it in prose, build the table:\n\n   1. Grep X's own identifier(s)\u002Fmarker(s) across the actual code: `grep -rn \"\u003CprecedentFn>\\|\u003CprecedentMarker>\" \u003Csrc>` (exclude test files). Integration patterns (snippet\u002Fscript injection, hooks, interceptors, lifecycle handlers) routinely span a **generation** step, a **preview\u002Ftransform** step, **and** a **publish\u002Frender-time** step — enumerate all of them; the publish\u002Frender-time site is the one a memory-driven spec most often drops.\n   2. Emit one row **per site X integrates at**:\n\n      | Precedent site (`file:line` + what runs there) | Spec coverage: Replicated \u002F Deferred-with-reason \u002F **OMITTED** |\n\n   3. Every grep hit MUST appear as a row. Decide each row's coverage by searching the spec for *that specific site* — never by trusting the spec's own prose summary that it \"follows X\".\n   4. **Every row marked OMITTED is its own Completeness Critical.** Do not fold multiple omitted sites into one finding, and **do not let a louder conflict elsewhere (e.g. a stale-anchor or deprecated-path finding) substitute for explicitly naming the omitted site** — an omission and a conflict are different defects and both must be reported. Each OMITTED Critical names the omitted site's `file:line` and the runtime path on which the feature fails.\n\n   Include the coverage table in the report, and feed each OMITTED row into the Completeness dimension of Phase 1 as a separate Critical.\n\n3.6. **UI\u002FInteraction Omission Check (MANDATORY omission check).** Step 3.5 targets precedent-site omissions; this step targets UI\u002Finteraction trigger surfaces with the same structure — a spec's own prose rarely states a QA expectation an experienced reviewer would apply by habit, so the check runs against a fixed lens table rather than the spec's own claims.\n\n   > **Action required**: Read [references\u002Fui-qa-lenses.md](references\u002Fui-qa-lenses.md) now for the canonical lens table before running this check.\n\n   For each lens row whose Trigger surface matches this spec, check whether the spec's Acceptance Criteria \u002F Functional Requirements \u002F Edge Cases satisfy that lens's Rule. A triggered lens the spec does not address is a **Completeness Warning** — cite the lens name and the nearest relevant spec section (or its absence). Feed each such row into the Completeness dimension of Phase 1 as its own finding (do not fold several lenses into one), and include them in the report. Skip any lens whose trigger surface the spec does not touch; do not force-fit lenses onto a non-UI spec.\n\n4. **Phase 1 — Parallel Dimension Review** — Use the Task tool to spawn 4 Sonnet subagents in parallel, one per review dimension. Pass each subagent the project context from Step 1 and the spec text from Step 2:\n\n   > **Action required**: Read [..\u002Freferences\u002Fsubagent-async-monitoring.md](..\u002Freferences\u002Fsubagent-async-monitoring.md) now — these four dimension subagents are one batch under that contract (distinct `name` per dimension, full-roster reconciliation on every signal, turn-boundary waiting, bounded escalation). Apply it before aggregating the dimension returns. **Release (mechanical gate — do not skip)**: named dispatches stay `idle` after finishing instead of exiting. After all four dimension returns are aggregated, call `TaskStop` once per dimension subagent with its `name` as `task_id` (current batch only; skip any still `running`). See the \"Releasing Completed Named Agents\" section of the reference.\n\n   | Subagent | Model | Dimension | Focus |\n   |---|---|---|---|\n   | Completeness | sonnet | Completeness | Missing required items, edge cases, pagination. For a **duplicate-sensitive write flow** (stock\u002Fbalance\u002Forder\u002Fpayment\u002Fprovisioning\u002Fquota) also check whether the spec specifies: (a) concurrent-request behavior, (b) the transaction\u002Fconsistency boundary spanning multiple DB writes, (c) handling of duplicate client retry \u002F double-click, (d) the failure response\u002Fstatus when stock\u002Fbalance\u002Fquota is exhausted or a lock\u002Fversion mismatch occurs. Missing any → a Completeness finding: **Critical** when the gap can cause double charge \u002F oversell \u002F lost ledger \u002F duplicate provisioning, otherwise **Warning** |\n   | Consistency | sonnet | Consistency | Terminology mismatches; Synonyms-to-Avoid violations |\n   | Feasibility | sonnet | Feasibility + Simplicity | Implementable with current stack and dependencies; also flag over-engineering — abstraction, configurability, or generality the stated scope does not yet require (surface as Warning). Check command-bearing AC with the AC baseline check (see the callout below the table). |\n   | Code Compatibility | sonnet | Code Compatibility | Conflicts with existing schema, API routes, type definitions |\n\n   > **Action required**: Read [references\u002Fac-baseline-check.md](references\u002Fac-baseline-check.md) now and pass its content to the Feasibility subagent — it defines the path exclusions, the Glob-only existence check, the single Grep-tool-call re-run and the `Baseline:` comparison.\n\n   Each subagent returns:\n   - **Confirmed findings** — dimension label, severity (Critical \u002F Warning \u002F Suggestion), file:line, description, improvement\n   - **Advisor candidates** — findings where two reasonable interpretations exist (include the specific choice and its consequence, ≤100 lines per candidate)\n\n   **Step 4b — Aggregate**: Combine and deduplicate findings by `{file}:{line}`. Cap advisor candidates at `advisor_budget` (default: 2), prioritizing Critical over Warning. Log any dropped candidates in the report.\n\n   The 4-dimension analysis runs the same way regardless of `--mode`; the mode only changes how findings are presented in step 5.\n\n4c. **Cross-Artifact Consistency (Analyze) — only when `--tasks \u003Cdir>` is set.** Steps 1–4 validate the spec in isolation. When a task directory already exists (re-validation after `ywc-task-generator`, or inside a convergence loop), verify the spec and tasks have not drifted apart — a defect class that spec-internal review structurally cannot see (an uncovered requirement makes no claim, so nothing conflicts, yet the feature ships incomplete). Build two tables before writing the report:\n\n   **Requirement Coverage** — one row per spec Functional Requirement \u002F Acceptance Criterion. Grep the task directory for each FR\u002FAC:\n\n   | Spec FR\u002FAC (`file:line`) | Covering task(s) (`tasks\u002F\u003Cid>`) | Replicated \u002F Deferred-with-reason \u002F **UNCOVERED** |\n\n   Every **UNCOVERED** row is its own Completeness Critical (a requirement no task implements).\n\n   **Task Provenance** — one row per task. Trace each task back to the spec:\n\n   | Task (`tasks\u002F\u003Cid>`) | Spec basis (`file:line` FR\u002FAC or named precedent) \u002F **ORPHAN** |\n\n   Every **ORPHAN** row (a task with no spec basis) is a Consistency Warning (unspecified scope). Then confirm the task dependency graph respects the spec's declared phase\u002Fsequence ordering — a task depending on a later-phase task is a Consistency finding.\n\n   Feed UNCOVERED rows into the Completeness dimension and ORPHAN \u002F dependency-order rows into the Consistency dimension as separate findings (do not fold multiple into one). Include both tables in the report.\n\n5. **Output Report — Branch on `--mode`**\n   - `standard` (default): emit the severity-classified finding report in the format below.\n   - `socratic`: emit a learning-question list per [references\u002Fsocratic-mode.md](references\u002Fsocratic-mode.md). Findings still drive question selection — they are just reshaped from \"fix X\" into \"what would happen if X?\". Completion Status is `SOCRATIC` and the gate score is informational only.\n\n## Review Dimensions\n\n| Dimension | Review Focus |\n|-----------|-------------|\n| Completeness | Missing required items (error handling, edge cases, pagination, etc.) |\n| Consistency | Terminology, format, or data structure mismatches across documents; spec terms appearing in the \"Synonyms to Avoid\" column of `docs\u002Fubiquitous-language.md` |\n| Feasibility | Whether the spec can be implemented with the current stack, and whether its command-bearing AC pass the [AC baseline check](references\u002Fac-baseline-check.md) |\n| Code Compatibility | Conflicts with existing DB schema, API route patterns |\n| Simplicity | Whether the spec specifies abstraction, configurability, or generality the stated scope does not yet require (over-engineering). Assessed within the Feasibility pass — surface speculative scope as a Warning, not a separate subagent. |\n\nUI\u002FInteraction omission mapping (Step 3.6):\n- **Completeness**: the spec triggers ≥1 lens from [references\u002Fui-qa-lenses.md](references\u002Fui-qa-lenses.md) but its Acceptance Criteria \u002F Functional Requirements \u002F Edge Cases do not satisfy that lens's rule.\n\n## Output Format\n\n```text\n## Spec Review Result: {filename}\n\n### Summary\n- Critical: N, Warning: M, Suggestion: K\n- Phase 2 advisor calls used: X of N ({single-advisor|council|none})   (N = applied budget; default 2, or the `--advisor-budget` value)\n\n### Precedent Site Coverage\n| Precedent site (`file:line` + what runs there) | Spec coverage: Replicated \u002F Deferred-with-reason \u002F OMITTED |\n|---|---|\n| ... | ... |\n\n### Critical Issues\n1. [{file}:{line}] [P1|P2] Description — Suggested improvement\n   (if P2) Advisor rationale: {one-line rationale}\n\n### Warning Issues\n1. [{file}:{line}] [P1|P2] Description — Suggested improvement\n\n### Suggestions\n1. [{file}:{line}] Description — Suggested improvement\n\n### Completion Status\n(One of: DONE | DONE_WITH_CONCERNS | BLOCKED | NEEDS_CONTEXT)\n```\n\n**Completion Status rules:**\n\nStatuses decided by the Confidence Gate result (`DONE`, `DONE_WITH_CONCERNS`, and the gate's `NEEDS_CONTEXT` \u002F `BLOCKED` bands) are defined only in the Band-to-status mapping table under [Confidence Gate](#confidence-gate) — that table is the single source; do not restate it here. The table below lists the pre-gate statuses, which are decided before (or instead of) the gate runs:\n\n| Pre-gate status | When to use |\n|--------|------------|\n| `BLOCKED` | Review cannot proceed — spec file missing, or a Phase 2 advisor returned an inconclusive verdict |\n| `NEEDS_CONTEXT` | `--spec` argument is missing or the file is empty\u002Funreadable |\n| `SOCRATIC` | `--mode socratic` was used; output is a learning-question list, not a gate verdict. Downstream skills (especially `ywc-task-generator`) must not consume this status as a handoff signal. |\n\n**Precedence — read this before emitting a status.** The Pre-gate status table above and the\nConfidence Gate's Band-to-status mapping are **not two independent rules**; the gate decision\nstatus comes only from the Band-to-status table. A spec with zero Critical findings scoring 78 is\n`NEEDS_CONTEXT` by the band, never `DONE`. Resolve it this way:\n\n1. **The Confidence Gate band decides the status.** It is the outer gate; run it last and let it win.\n2. **The finding count only splits the PROCEED band** — `DONE` when no Critical findings remain,\n   `DONE_WITH_CONCERNS` when any do. Below PROCEED the finding count does not upgrade the status:\n   a zero-Critical spec in REVIEW is `NEEDS_CONTEXT`, and in STOP it is `BLOCKED`.\n3. **Input failures short-circuit both.** A missing\u002Fempty\u002Funreadable `--spec` is `NEEDS_CONTEXT`, and\n   a missing spec file or an inconclusive Phase 2 advisor is `BLOCKED` — emitted *without* computing\n   a gate score, because there is nothing to score.\n\nSo `NEEDS_CONTEXT` and `BLOCKED` each have two legitimate triggers: an input failure (rule 3) or a\ngate band (rule 1). That overlap is intentional, not a conflict — but always say which one fired, so\na caller can tell \"I could not read your spec\" from \"I read it and it is not ready\".\n\n`SOCRATIC` sits outside this hierarchy: it is a mode, not a verdict, and no gate band produces it.\n\n**Re-plan handoff on `DONE_WITH_CONCERNS`** — after printing the report body, append the following one-liner so the user (or an interactive orchestrator) can resolve findings via `ywc-plan` Re-plan Mode instead of rewriting the spec from scratch:\n\n```\nTo address the Critical findings above without losing validated sections, run:\n\u002Fywc-plan --update-spec \u003Cspec-path> --failure-context \"\u003Cone-paragraph summary of the Critical findings>\"\n```\n\nThis routes findings into the `## Iteration N Amendments` append flow described in `ywc-plan` Step 4c, preserving validated portions of the spec. Print this line only on `DONE_WITH_CONCERNS` — `DONE`, `BLOCKED`, `NEEDS_CONTEXT`, and `SOCRATIC` do not benefit from Re-plan Mode. For programmatic consumers (e.g., `ywc-agentic`), the Programmatic Consumer Policy below already invokes the equivalent flow; this surface is for interactive users who would otherwise re-write from scratch.\n\n> **HTML mode (`--format html`)** — emits the same findings as a self-contained HTML report: severity color coding, tab navigation, and a `Copy as Markdown` button. Structure and conventions follow [html-output.md](..\u002Freferences\u002Fhtml-output.md). The Markdown surface is preserved inside the file, so downstream integration is unaffected.\n\n## Advisor Escalation Policy\n\nThis skill runs Phase 1 as 4 parallel Sonnet subagents (one per dimension) and aggregates their findings. For a small number of **genuinely ambiguous** findings from Phase 1, the orchestrator escalates to an Opus advisor using the Task tool with `model: opus`. This follows **Pattern B** from [advisor-pattern.md](..\u002Freferences\u002Fadvisor-pattern.md) — parallel executors in Phase 1, frontier judgment applied only where it actually matters in Phase 2.\n\n**Budget**: up to 2 Opus advisor calls per invocation by default. Unused budget is good. Spec review is smaller in scope than impl-review, so the cap is tighter.\n\n**Complexity-Aware Budget Defaults** — right-size the advisor\u002Fcouncil spend to the spec's scope *before* Phase 1 (an explicit `--advisor-budget` always overrides this default):\n\n| Spec scope | Default advisor budget |\n|---|---|\n| Single-section follow-up \u002F ≤2 modules touched | 0–1 |\n| Standard spec \u002F 3–5 modules | 2 (current default) |\n| Cross-cutting \u002F ≥15 acceptance criteria or >5 modules | 2, and prefer Council Escalation for the high-stakes finding |\n\nThe Sonnet 4-dimension fan-out (Phase 1) always runs regardless of scope — it is cheap and catches omissions. Only the Opus Phase 2 \u002F Council spend is right-sized here, so the \"Right Job, Right Tool\" saving never comes at the cost of missed findings.\n\n**`--advisor-budget \u003Cn>` override**: when set, `\u003Cn>` is the hard ceiling on Phase 2 Opus calls for this invocation (default 2). The report header's `Phase 2 advisor calls used: X of N` line reflects the applied budget as `N` so an orchestrator can track cumulative spend across iterations. When `--advisor-budget 0`, Phase 2 escalation is **disabled entirely**: each finding that would have escalated is instead reported as a normal Suggestion carrying a one-line `would have escalated but budget=0` rationale — it is **never silently dropped**. This lets a loop driver (e.g. `ywc-spec-ready`) run cheap deterministic passes and spend advisor budget only when it chooses to.\n\n**Escalation conditions** — each must satisfy all three properties from [advisor-pattern.md §5](..\u002Freferences\u002Fadvisor-pattern.md) (objective trigger, irreversibility, ambiguity):\n\n1. **Feasibility uncertainty** — The spec appears feasible but hinges on an assumption about an undocumented library feature, version-specific behavior, or performance characteristic. A wrong assumption would make the spec unimplementable or force mid-implementation scope cuts.\n2. **Code Compatibility conflict with plausible-either-way resolution** — The spec conflicts with existing code (schema, API contract, or type definition), and both \"spec adapts to code\" and \"code refactors to spec\" have plausible precedent in the project. The advisor picks which side moves.\n3. **Completeness borderline** — The spec omits a feature that could be intentional (out of scope, deferred to another spec) or an oversight. The wrong verdict delays implementation (if flagged as gap when actually out of scope) or generates incorrect downstream tasks (if dismissed when actually a gap).\n4. **Cross-document consistency conflict** — The spec conflicts with another spec or architecture document in the project, and resolving the conflict requires domain judgment about which document is the current source of truth.\n\n**Context payload rules** (critical for cost discipline):\n- Forward only the decision: the specific spec clause, the conflicting code or spec excerpt, and the constraint (≤100 lines total).\n- Do NOT forward: the full spec, the full target code, or the CLAUDE.md contents.\n- The advisor returns a short verdict (≤200 words) containing the recommended resolution and a one-line rationale.\n\n**Non-goals** — do not escalate for these:\n- Terminology nitpicks or formatting inconsistencies (these are Phase 1 Warnings, not advisor material).\n- Generic \"the spec could be clearer\" observations without a specific failure mode.\n- Stylistic preferences about how requirements are expressed.\n- Missing sections that are clearly an author oversight (just report it as a Warning).\n\nReport escalations in the output: mark Phase 2 findings with `[P2]` prefix and include the advisor's one-line rationale. This preserves auditability of which findings involved frontier judgment.\n\n## Council Escalation (High-Stakes Trade-offs)\n\nWhen a finding involves a genuine architectural trade-off — where two reasonable positions exist and the wrong choice has significant implementation cost — the single Opus advisor may be replaced by a 4-voice **Council** review. Use council escalation when all three conditions hold:\n\n1. The trade-off has long-term architectural consequences (e.g., sync vs. async boundary, data ownership model, API versioning strategy).\n2. The single-advisor verdict would likely depend on unstated assumptions about team preferences or future requirements.\n3. The finding is Critical or Warning severity — not a Suggestion.\n\n**Council voices** (spawn one Opus subagent per voice, passing only the specific trade-off excerpt — not the full spec):\n\n| Voice | Role | Anti-bias Mandate |\n|-------|------|-------------------|\n| Architect | Correctness and long-term maintainability | Challenge convenience-driven choices |\n| Skeptic | Break the dominant assumption; find what everyone is taking for granted | Ignore sunk cost arguments |\n| Pragmatist | Shipping speed and team execution risk | Ignore theoretical purity |\n| Critic | Failure modes; what breaks at scale or under adversarial conditions | Assume the happy path is wrong |\n\n**Anti-anchoring rule**: Each voice receives only the trade-off description — not the other voices' verdicts. Collect all four independently before synthesizing.\n\n> **Action required**: Read [..\u002Freferences\u002Fsubagent-async-monitoring.md](..\u002Freferences\u002Fsubagent-async-monitoring.md) now — the four Council voices are one batch under that contract (distinct `name` per voice, full-roster reconciliation on every signal, turn-boundary waiting, bounded escalation). **Release (mechanical gate — do not skip)**: named dispatches stay `idle` after finishing instead of exiting. After all four voice verdicts are collected and synthesized, call `TaskStop` once per voice subagent with its `name` as `task_id` (current batch only; skip any still `running`). See the \"Releasing Completed Named Agents\" section of the reference.\n\n**Council output format** (append to the relevant finding in the report):\n```\n[COUNCIL] {trade-off description}\n- Architect: {verdict, ≤2 sentences}\n- Skeptic: {verdict, ≤2 sentences}\n- Pragmatist: {verdict, ≤2 sentences}\n- Critic: {verdict, ≤2 sentences}\nConsensus: {agreed recommendation, or \"No consensus — strongest dissent: \u003Cvoice>: \u003Creason>\"}\n```\n\n### Profile Override (Focus-Specific Voices)\n\nWhen `--focus \u003Carea>` is set on the invocation and Council Escalation triggers, replace the 4 generic voices above with 3–4 domain expert profiles from [references\u002Fexpert-profiles.md](references\u002Fexpert-profiles.md). The Anti-anchoring rule, output format, and budget rule all apply unchanged — only the voice labels and heuristics change.\n\n| `--focus` value | Profiles to use (in order) |\n|---|---|\n| `requirements` | Requirements Quality + Specification by Example + Use Case Modeling |\n| `architecture` | Interface Design + Service Boundaries + Integration Patterns + Production Resilience |\n| `testing` | Risk-Based Testing + Three Amigos + Quality Attribute |\n| `compliance` | Production Resilience + Cloud-Native Operations + Audit Trail |\n\nUse generic Council (no `--focus`) when the trade-off spans two or more areas, or when the spec covers multiple domains and no single focus dominates. Mixing generic voices and focus profiles within a single Council session breaks anti-anchoring — pick one mode per session.\n\nCouncil escalation consumes the full 2-advisor budget. Do not use council and single-advisor escalation in the same invocation.\n\n## Confidence Gate\n\nThis skill applies the [Confidence Gate](..\u002Freferences\u002Fconfidence-gate.md) before declaring the spec ready for `task-generator`.\n\n**Required dimensions** (must each score ≥ 70):\n\n- **Scope clarity** — A spec with undefined scope cannot be decomposed into tasks. If scope is fuzzy, the gate forces STOP.\n- **Root cause identified** — The user need behind the spec must be named. Surface-level requirements without a stated user need produce tasks that solve the wrong problem.\n\n**Band-to-status mapping** for this skill (conforms to [confidence-gate.md §5](..\u002Freferences\u002Fconfidence-gate.md) — PROCEED splits on Critical findings so `DONE_WITH_CONCERNS` stays reachable):\n\n| Gate band | Completion status | Action |\n|-----------|-------------------|--------|\n| PROCEED (≥ 90), no Critical findings | DONE | Spec is ready for `task-generator`. |\n| PROCEED (≥ 90), Critical findings present | DONE_WITH_CONCERNS | Resolve the Critical findings via the Re-plan handoff before `task-generator`. |\n| REVIEW (70 – 89) | NEEDS_CONTEXT | Present 1-3 open questions; consider council escalation if multiple dimensions are \u003C 80. |\n| STOP (\u003C 70) | BLOCKED | Report which dimensions are weak; do not hand off to `task-generator`. |\n\nThe gate score must appear in the report header alongside the Critical\u002FWarning\u002FSuggestion finding counts. A required dimension scoring below 50 forces STOP regardless of aggregate.\n\n## Integration\n\n- **Primary upstream** — `ywc-spec-writer`: validates `docs\u002Fspecification\u002F` section files (`01-overview.md` … `07-glossary.md`). All 4 review dimensions (completeness, consistency, feasibility, code compatibility) apply fully.\n- **Secondary upstream** — `ywc-plan`: can validate feature plan documents in `docs\u002Fywc-plans\u002F`. Completeness and consistency dimensions apply; feasibility and code compatibility apply partially because plan documents are less implementation-specific.\n- **Cross-artifact (Analyze)** — `ywc-task-generator` output (`tasks\u002F`): once tasks exist, pass the directory via `--tasks \u003Cdir>` (alongside `--spec`) to run the Cross-Artifact Consistency pass (Step 4c) — confirms every spec requirement is covered by a task and no task is orphaned. Passing a task directory as `--spec` is still a misuse; `--tasks` is the correct channel.\n- **Downstream**: `ywc-task-generator` — hand off only when completion status is `DONE`. `DONE_WITH_CONCERNS` requires spec revision first.\n\n### Programmatic Consumer Policy\n\nWhen consumed by an orchestrator that cannot prompt for human input (e.g., ywc-agentic):\n\n| Completion Status | Action |\n| --- | --- |\n| DONE | Proceed to ywc-task-generator |\n| DONE_WITH_CONCERNS | Re-invoke ywc-plan with --update-spec + --failure-context + --non-interactive (max 1 retry), then re-validate |\n| BLOCKED | Stop execution; surface a structured triage report containing: (1) attempted triage steps, (2) verbatim blocker text, (3) proposed default action |\n| NEEDS_CONTEXT | Stop execution; surface a structured triage report containing: (1) attempted triage steps, (2) verbatim blocker text, (3) proposed default action |\n| SOCRATIC | Stop execution and report to user — SOCRATIC output is not a handoff signal; re-run without `--mode socratic` to obtain a gate verdict |\n\n> **Looped consumer**: the table's `max 1 retry` is the default for non-loop consumers (e.g. `ywc-agentic` Step 3 historically). When **multi-iteration** convergence is required, the dedicated `ywc-spec-ready` skill drives the `DONE_WITH_CONCERNS → ywc-plan --update-spec → re-validate` cycle repeatedly under its own `--max-iterations` cap and stall guards, and passes the remaining advisor budget via `--advisor-budget`. Use `ywc-spec-ready` instead of hand-rolling more than one retry.\n",{},"main","7e8ac872e53df8c89d75663e031626281a2a2f5c","claude-code\u002Fskills\u002Fywc-spec-validate",1,"https:\u002F\u002Fgithub.com\u002Fyongwoon\u002Fywc-agent-toolkit","https:\u002F\u002Fgithub.com\u002Fyongwoon","https:\u002F\u002Fwww.webhani.com","2026-11-08 17:14:59",{"source":7,"skillName":5,"commands":31,"prompt":41,"zipAvailable":42},[32,35,38],{"tool":33,"command":34},"npx","npx skills add yongwoon\u002Fywc-agent-toolkit --skill ywc-spec-validate",{"tool":36,"command":37},"bunx","bunx skills add yongwoon\u002Fywc-agent-toolkit --skill ywc-spec-validate",{"tool":39,"command":40},"pnpm","pnpm dlx skills add yongwoon\u002Fywc-agent-toolkit --skill ywc-spec-validate","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fyongwoon\u002Fywc-agent-toolkit\u002Fblob\u002Fmain\u002Fclaude-code\u002Fskills\u002Fywc-spec-validate\u002FSKILL.md",true,[44,50,54,58,62,65,69,73,77,81],{"path":45,"resourceType":46,"resourceTypeDesc":47,"sizeBytes":48,"sourceHtmlUrl":49},"README.en.md","REFERENCE","参考文档",2196,"https:\u002F\u002Fgithub.com\u002Fyongwoon\u002Fywc-agent-toolkit\u002Fblob\u002Fmain\u002Fclaude-code\u002Fskills\u002Fywc-spec-validate\u002FREADME.en.md",{"path":51,"resourceType":46,"resourceTypeDesc":47,"sizeBytes":52,"sourceHtmlUrl":53},"README.es.md",3300,"https:\u002F\u002Fgithub.com\u002Fyongwoon\u002Fywc-agent-toolkit\u002Fblob\u002Fmain\u002Fclaude-code\u002Fskills\u002Fywc-spec-validate\u002FREADME.es.md",{"path":55,"resourceType":46,"resourceTypeDesc":47,"sizeBytes":56,"sourceHtmlUrl":57},"README.ja.md",2391,"https:\u002F\u002Fgithub.com\u002Fyongwoon\u002Fywc-agent-toolkit\u002Fblob\u002Fmain\u002Fclaude-code\u002Fskills\u002Fywc-spec-validate\u002FREADME.ja.md",{"path":59,"resourceType":46,"resourceTypeDesc":47,"sizeBytes":60,"sourceHtmlUrl":61},"README.ko.md",2280,"https:\u002F\u002Fgithub.com\u002Fyongwoon\u002Fywc-agent-toolkit\u002Fblob\u002Fmain\u002Fclaude-code\u002Fskills\u002Fywc-spec-validate\u002FREADME.ko.md",{"path":63,"resourceType":46,"resourceTypeDesc":47,"sizeBytes":60,"sourceHtmlUrl":64},"README.md","https:\u002F\u002Fgithub.com\u002Fyongwoon\u002Fywc-agent-toolkit\u002Fblob\u002Fmain\u002Fclaude-code\u002Fskills\u002Fywc-spec-validate\u002FREADME.md",{"path":66,"resourceType":46,"resourceTypeDesc":47,"sizeBytes":67,"sourceHtmlUrl":68},"README.zh.md",2991,"https:\u002F\u002Fgithub.com\u002Fyongwoon\u002Fywc-agent-toolkit\u002Fblob\u002Fmain\u002Fclaude-code\u002Fskills\u002Fywc-spec-validate\u002FREADME.zh.md",{"path":70,"resourceType":46,"resourceTypeDesc":47,"sizeBytes":71,"sourceHtmlUrl":72},"references\u002Fac-baseline-check.md",2841,"https:\u002F\u002Fgithub.com\u002Fyongwoon\u002Fywc-agent-toolkit\u002Fblob\u002Fmain\u002Fclaude-code\u002Fskills\u002Fywc-spec-validate\u002Freferences\u002Fac-baseline-check.md",{"path":74,"resourceType":46,"resourceTypeDesc":47,"sizeBytes":75,"sourceHtmlUrl":76},"references\u002Fexpert-profiles.md",9135,"https:\u002F\u002Fgithub.com\u002Fyongwoon\u002Fywc-agent-toolkit\u002Fblob\u002Fmain\u002Fclaude-code\u002Fskills\u002Fywc-spec-validate\u002Freferences\u002Fexpert-profiles.md",{"path":78,"resourceType":46,"resourceTypeDesc":47,"sizeBytes":79,"sourceHtmlUrl":80},"references\u002Fsocratic-mode.md",6186,"https:\u002F\u002Fgithub.com\u002Fyongwoon\u002Fywc-agent-toolkit\u002Fblob\u002Fmain\u002Fclaude-code\u002Fskills\u002Fywc-spec-validate\u002Freferences\u002Fsocratic-mode.md",{"path":82,"resourceType":46,"resourceTypeDesc":47,"sizeBytes":83,"sourceHtmlUrl":84},"references\u002Fui-qa-lenses.md",3425,"https:\u002F\u002Fgithub.com\u002Fyongwoon\u002Fywc-agent-toolkit\u002Fblob\u002Fmain\u002Fclaude-code\u002Fskills\u002Fywc-spec-validate\u002Freferences\u002Fui-qa-lenses.md",[86,92,98,105,112,119],{"skillNo":87,"name":88,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":8,"repositoryNo":11,"starsCount":12,"licenseSpdxId":13,"licenseFileUrl":6,"sourceHtmlUrl":89,"categories":90,"keywords":91,"publishedAt":17},"01M72TQ84GTCPCV2VJ778V17TW","missing-frontmatter","https:\u002F\u002Fgithub.com\u002Fyongwoon\u002Fywc-agent-toolkit\u002Fblob\u002Fmain\u002Fscripts\u002Ffixtures\u002Fcodex-description-limit\u002Fmissing-frontmatter\u002FSKILL.md",[],[],{"skillNo":93,"name":94,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":8,"repositoryNo":11,"starsCount":12,"licenseSpdxId":13,"licenseFileUrl":6,"sourceHtmlUrl":95,"categories":96,"keywords":97,"publishedAt":17},"01M72TQ6M4BFWE7P9EPW0GZDCM","malformed-frontmatter","https:\u002F\u002Fgithub.com\u002Fyongwoon\u002Fywc-agent-toolkit\u002Fblob\u002Fmain\u002Fscripts\u002Ffixtures\u002Fcodex-description-limit\u002Fmalformed-frontmatter\u002FSKILL.md",[],[],{"skillNo":99,"name":100,"nameZh":6,"description":101,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":8,"repositoryNo":11,"starsCount":12,"licenseSpdxId":13,"licenseFileUrl":6,"sourceHtmlUrl":102,"categories":103,"keywords":104,"publishedAt":17},"01M72TQ3ESGKXZ4QRWGTX92BKJ","fixture-501-fail","(ywc) Use when aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa Do not use for x.","https:\u002F\u002Fgithub.com\u002Fyongwoon\u002Fywc-agent-toolkit\u002Fblob\u002Fmain\u002Fscripts\u002Ffixtures\u002Fcodex-description-limit\u002F501-fail\u002FSKILL.md",[],[],{"skillNo":106,"name":107,"nameZh":6,"description":108,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":8,"repositoryNo":11,"starsCount":12,"licenseSpdxId":13,"licenseFileUrl":6,"sourceHtmlUrl":109,"categories":110,"keywords":111,"publishedAt":17},"01M72TQ0HJ8H3R37J9WPWGKJZR","fixture-500-pass","(ywc) Use when aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa Do not use for x.","https:\u002F\u002Fgithub.com\u002Fyongwoon\u002Fywc-agent-toolkit\u002Fblob\u002Fmain\u002Fscripts\u002Ffixtures\u002Fcodex-description-limit\u002F500-pass\u002FSKILL.md",[],[],{"skillNo":113,"name":114,"nameZh":6,"description":115,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":8,"repositoryNo":11,"starsCount":12,"licenseSpdxId":13,"licenseFileUrl":6,"sourceHtmlUrl":116,"categories":117,"keywords":118,"publishedAt":17},"01M72TPWNCC7NEYD0D9N12KSQS","ywc-worktrees","(ywc) Use when creating, auditing, pruning, or resolving git worktrees for isolated task workspaces, whether delegated from an executor or used standalone. Triggers: \"worktree 생성\", \"worktree audit\", \"worktree 정리\", \"worktree prune\", \"ywc-worktrees\", \"create worktree\". Do not use for branch creation\u002Fdeletion alone, PR creation (use ywc-create-pr), CI or merge delivery (use ywc-finish-branch), task orchestration (use ywc-sequential-executor or ywc-parallel-executor), or non-git isolation.","https:\u002F\u002Fgithub.com\u002Fyongwoon\u002Fywc-agent-toolkit\u002Fblob\u002Fmain\u002Fplugins\u002Fywc-agent-toolkit\u002Fskills\u002Fywc-worktrees\u002FSKILL.md",[],[],{"skillNo":120,"name":121,"nameZh":6,"description":122,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":8,"repositoryNo":11,"starsCount":12,"licenseSpdxId":13,"licenseFileUrl":6,"sourceHtmlUrl":123,"categories":124,"keywords":125,"publishedAt":17},"01M72TPRNR3E6QWB91GM98ABBV","ywc-wayfinder","(ywc) Use when a large or uncertain change needs multi-session discovery with a deterministic local map and exactly one active ticket before planning or implementation. Triggers: \"wayfinder\", \"길찾기\", \"discovery map\", \"multi-session discovery\", \"탐색 ticket\", \"経路整理\", \"探索マップ\", \"ywc-wayfinder\". Do not use for ordinary planning (use ywc-plan), idea clarification (use ywc-brainstorm), direct implementation, or tracker\u002Fdatabase writes outside the local Markdown map.","https:\u002F\u002Fgithub.com\u002Fyongwoon\u002Fywc-agent-toolkit\u002Fblob\u002Fmain\u002Fplugins\u002Fywc-agent-toolkit\u002Fskills\u002Fywc-wayfinder\u002FSKILL.md",[],[]]