[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7CS0Z5SF91WGW5SPP168FPX":3,"skill-same:01M7CS0Z5SF91WGW5SPP168FPX":61},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":24,"commitSha":25,"skillDirectory":26,"forksCount":27,"repositoryHtmlUrl":28,"repositoryHomepageUrl":29,"ownerHtmlUrl":30,"ownerBlogUrl":6,"ownerDescription":31,"updatedAt":32,"install":33,"resources":46},"01M7CS0Z5SF91WGW5SPP168FPX","skill-pipeline",null,"Pipeline orchestrator that classifies incoming coding tasks and routes them through the correct combination of skills at the right depth. Implements two feedback loops: the inner loop (detect, verify, recover) runs within a session via plan-interview, intent-framed-agent, context-surfing, verify-gate, simplify-and-harden, and self-improvement. The outer loop (inspect, encode, regress-test) runs across sessions via learning-aggregator, harness-updater, and eval-creator. pre-flight-check bridges the two by surfacing accumulated knowledge at session start. Handles standard, team-based, CI, and outer-loop pipeline variants. Does not replace individual skills; dispatches to them.","YYC-Cube\u002FYYC3-Knowledge-Base","YYC-Cube","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F235006543?v=4","01M79J2PH4SWD2X0825X6PN748","yyc3","01M79J2PHVK3F6ZBYT8K2SCZCW",1,"MIT","https:\u002F\u002Fgithub.com\u002FYYC-Cube\u002FYYC3-Knowledge-Base\u002Fblob\u002Fmain\u002FYYC3-Skills-技能包\u002FYYC3-Skills-技能包-pskoett-ai-skills\u002Fskills\u002Fskill-pipeline\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: skill-pipeline\ndescription: >\n  Pipeline orchestrator that classifies incoming coding tasks and routes them\n  through the correct combination of skills at the right depth. Implements two\n  feedback loops: the inner loop (detect, verify, recover) runs within a session\n  via plan-interview, intent-framed-agent, context-surfing, verify-gate,\n  simplify-and-harden, and self-improvement. The outer loop (inspect, encode,\n  regress-test) runs across sessions via learning-aggregator, harness-updater,\n  and eval-creator. pre-flight-check bridges the two by surfacing accumulated\n  knowledge at session start. Handles standard, team-based, CI, and outer-loop\n  pipeline variants. Does not replace individual skills; dispatches to them.\n---\n\n# Skill Pipeline\n\nThe conductor, not a player. This skill classifies tasks, selects the pipeline variant, calibrates depth, and orchestrates handoffs between skills. It produces no artifacts of its own — its output is routing decisions that activate other skills.\n\n## Task Classification\n\nOn every coding task, classify before acting. Evaluate scope signals and map to a task class.\n\n**Input signals:** file count, task description, existing plan\u002Fhandoff files, batch indicators, CI environment.\n\n```\nTask received\n  │\n  ├─ Trivial (typo, rename, version bump)\n  │  → No skills. Just do it.\n  │\n  ├─ Small (isolated fix, single-file, \u003C10 logic lines)\n  │  → verify-gate + simplify-and-harden\n  │\n  ├─ Medium (feature in known area, 2-5 files)\n  │  → intent-framed-agent + verify-gate + simplify-and-harden\n  │\n  ├─ Large (complex refactor, new architecture, unfamiliar codebase, high-risk logic)\n  │  → Full standard pipeline\n  │  → Recommend \u002Fplan-interview before starting\n  │\n  ├─ Long-running (multi-session, high context pressure, prior handoff exists)\n  │  → Full standard pipeline with context-surfing as critical skill\n  │\n  └─ Batch (multiple features from spec, 5+ discrete tasks, issue triage)\n     → Team-based pipeline (agent-teams-simplify-and-harden)\n```\n\nWhen uncertain, start with Medium. Add skills if drift or quality issues appear mid-task.\n\nFor detailed heuristics, edge cases, and examples: read `references\u002Fclassification-rules.md`.\n\n## Pipeline Selection\n\nRoute task class to the right variant:\n\n| Task Class | Variant | Rationale |\n|------------|---------|-----------|\n| Trivial | None | No overhead needed |\n| Small | Standard (minimal) | Verify + S&H only |\n| Medium | Standard (partial) | Scope monitoring + verify + review |\n| Large | Standard (full) | Full inner loop with planning |\n| Long-running | Standard (full) | Context-surfing is critical |\n| Batch | Team-based | Breadth over depth |\n| CI environment | CI | Headless review |\n| Periodic | Outer loop | Cross-session improvement |\n\n**Heuristic:** Standard pipeline for **depth** (single complex feature). Team-based pipeline for **breadth** (batch of tasks). CI pipeline when `CI=true` or `GITHUB_ACTIONS=true`.\n\n## Activation Sequences\n\n### Standard Pipeline (Inner Loop)\n\n```\npre-flight-check (SessionStart hook — surfaces prior learnings)\n  → classify\n  → (recommend \u002Fplan-interview if Large or Long-running)\n  → intent-framed-agent (at planning-to-execution transition)\n  → context-surfing (auto-activates when intent frame + plan exist; concurrent with intent monitoring)\n  → [IMPLEMENTATION]\n  → verify-gate (compile + test + lint; fix loop if red)\n  → simplify-and-harden (post-completion, if non-trivial diff)\n  → self-improvement (on errors, corrections, or S&H learning candidates)\n```\n\n**Skill-by-class activation:**\n\n| Skill | Trivial | Small | Medium | Large | Long-running |\n|-------|---------|-------|--------|-------|-------------|\n| pre-flight-check | Hook | Hook | Hook | Hook | Hook |\n| plan-interview | - | - | - | Recommend | Recommend |\n| intent-framed-agent | - | - | Activate | Activate | Activate |\n| context-surfing | - | - | - | Activate | Critical |\n| verify-gate | - | Activate | Activate | Activate | Activate |\n| simplify-and-harden | - | If non-trivial | If non-trivial | If non-trivial | If non-trivial |\n| self-improvement | On error only | On error only | On error\u002Fcompletion | On error\u002Fcompletion | On error\u002Fcompletion |\n\n### Team-Based Pipeline\n\n```\nclassify (Batch)\n  → (recommend \u002Fplan-interview if no spec exists)\n  → agent-teams-simplify-and-harden\n    ├─ Team lead emits Intent Frame #1\n    ├─ Phase 1: parallel implementation agents\n    ├─ verify-gate (compile + test + lint)\n    ├─ Phase 2: parallel audit agents (simplify, harden, spec)\n    ├─ Fix loop (up to 3 audit rounds)\n    └─ Learning loop output\n  → self-improvement\n```\n\n### CI Pipeline\n\n```\nclassify (CI detected)\n  → simplify-and-harden-ci (headless scan, PR changed files only)\n  → self-improvement-ci (pattern aggregation, promotion recommendations)\n```\n\n### Outer Loop Pipeline\n\nThe outer loop runs across sessions, not within them. Trigger on cadence (weekly, sprint boundary) or when `pre-flight-check` surfaces promotion-ready patterns.\n\n```\nlearning-aggregator (read .learnings\u002F, find patterns, rank promotion candidates)\n  → harness-updater agent (apply promotions to CLAUDE.md, AGENTS.md, copilot-instructions.md)\n  → eval-creator (create permanent test cases from promoted patterns)\n  → eval-creator run (regression check on all existing evals)\n```\n\n**When to trigger the outer loop:**\n- Weekly: recommended minimum cadence\n- Sprint boundary: after a burst of sessions\n- When `pre-flight-check` reports promotion-ready count > 3\n- After a significant incident or recurring failure\n- Manually: user invokes `\u002Flearning-aggregator`\n\n**Outer loop is always human-gated.** `learning-aggregator` produces a gap report. `harness-updater` shows diffs for approval. No automatic writes to instruction files without human review.\n\n## Depth Calibration\n\nNot just which skills — how deep each goes:\n\n| Dimension | Small | Medium | Large | Long-running | Batch |\n|-----------|-------|--------|-------|-------------|-------|\n| Pre-flight check | Hook | Hook | Hook | Hook | Hook |\n| Planning passes | 0 | 0-1 | 1-2 | Deep iterative | Per-task or umbrella |\n| Intent frame | - | Single frame | Full frame + monitoring | Full + handoff | Team lead frame |\n| Context-surfing | - | - | Active | Critical (exit protocol ready) | Lightweight drift checks |\n| Verify-gate | Compile + test | Compile + test | Compile + test + lint | Compile + test + lint | Compile + test (per round) |\n| S&H budget | 20% diff, 60s | 20% diff, 60s | 20% diff, 60s | 20% diff, 60s | 30% team growth cap |\n| Audit rounds (teams) | - | - | - | - | Up to 3 |\n| Self-improvement | Error-triggered | Error-triggered | Error + S&H feed | Error + S&H feed | Error + teams feed |\n\n## Handoff Rules\n\nArtifacts flow between skills. The orchestrator ensures each skill receives what it needs.\n\n**Key handoffs:**\n\n1. **Plan file** (`docs\u002Fplans\u002Fplan-NNN-\u003Cslug>.md`) — produced by `plan-interview`, consumed by `intent-framed-agent` (context), `context-surfing` (wave anchor), `agent-teams` (task extraction).\n\n2. **Intent Frame** — produced by `intent-framed-agent`, consumed by `context-surfing` (wave anchor strengthening). Copied into handoff files on drift exit.\n\n3. **Handoff file** (`.context-surfing\u002Fhandoff-[slug]-[timestamp].md`) — produced by `context-surfing` on drift exit, consumed by next session for resume.\n\n4. **Verify-gate signal** — produced by `verify-gate` (pass\u002Ffail + diagnostics), consumed by `simplify-and-harden` (only activates after green gate) and fix loop (on failure).\n\n5. **Learning candidates** (`learning_loop.candidates`) — produced by `simplify-and-harden` and `agent-teams`, consumed by `self-improvement` for pattern tracking.\n\n6. **Learning entries** (`.learnings\u002F*.md`) — produced by `self-improvement`, consumed by `learning-aggregator` for cross-session analysis and by `pre-flight-check` at session start.\n\n7. **Gap report** — produced by `learning-aggregator`, consumed by `harness-updater` agent for promotion and `eval-creator` for test case generation.\n\n8. **Eval cases** (`.evals\u002Fcases\u002F*.md`) — produced by `eval-creator`, consumed by regression runs and surfaced by `pre-flight-check`.\n\n**Precedence:** If `context-surfing` and `intent-framed-agent` both fire simultaneously, context-surfing's exit takes precedence. Degraded context makes scope checks unreliable.\n\nFor the full artifact\u002Fsignal\u002Fbudget table: read `references\u002Fhandoff-matrix.md`.\n\n## Decision Points\n\nThe orchestrator intervenes at these moments:\n\n### Task Arrival\nClassify the task. Select pipeline variant and depth. Emit routing decision. If Large\u002FLong-running, recommend `\u002Fplan-interview`. If Batch, recommend team-based variant.\n\n### Plan Approval\nWhen user approves a plan from `plan-interview`, flow directly into the execution stage — no separate \"should I proceed?\" prompt. This means activating `intent-framed-agent` to emit an Intent Frame. The intent frame itself still requires user confirmation before coding begins (that confirmation is part of `intent-framed-agent`, not an extra gate). Populate task tracking with checklist items.\n\n### Planning-to-Execution Transition\nWhen no plan-interview was used and the user signals readiness (\"go ahead\", \"implement this\", \"let's start\"), activate `intent-framed-agent`. Emit Intent Frame. Wait for user confirmation of the frame before coding.\n\n### Implementation Complete\nActivate `verify-gate` to run compile, test, and lint checks. If any fail, enter the fix loop (up to 3 attempts per phase). Once all checks pass and the diff meets the non-trivial threshold (see `references\u002Fclassification-rules.md`), activate `simplify-and-harden`. If the diff is trivial, signal completion directly after verify-gate passes.\n\n### Drift Detected\nIf `context-surfing` fires a drift exit, stop execution. Write handoff file. If the task was classified below Large, consider re-classifying upward for the next session.\n\n### Session Resume\nCheck for handoff files in `.context-surfing\u002F`. If found, read completely. Re-establish context from handoff. Re-classify if needed. Resume from recommended re-entry point.\n\n## Overrides\n\nUsers can override any routing decision:\n\n- **Force depth:** `depth=small` \u002F `depth=large` — override classification\n- **Force variant:** `variant=teams` \u002F `variant=standard` — override pipeline selection\n- **Skip review:** `--no-review` — skip `simplify-and-harden`\n- **Force planning:** invoke `\u002Fplan-interview` on any task regardless of classification\n- **Skip all skills:** user says \"just do it\" on a non-trivial task — respect the override\n\n## Re-classification\n\nTasks can change class mid-execution. Watch for:\n\n- **Escalation signals:** scope expanded beyond original estimate, many more files affected than expected, `context-surfing` drift exit, `intent-framed-agent` detects significant scope change\n- **De-escalation signals:** task turns out simpler than planned, plan reveals minimal changes needed\n\nWhen re-classification is warranted:\n1. Note the signal that triggered re-classification\n2. Adjust active skills (add or remove pipeline stages)\n3. If escalating to Large: recommend `\u002Fplan-interview` if no plan exists\n4. If de-escalating: drop unnecessary stages, proceed with lighter pipeline\n\n## Anti-Patterns\n\n- **Do NOT re-implement skill logic.** This skill classifies and dispatches. Each individual skill owns its own procedure.\n- **Do NOT auto-invoke plan-interview.** It is a human gate. Recommend it; let the user decide.\n- **Do NOT create a third monitoring layer.** During execution, `intent-framed-agent` monitors scope and `context-surfing` monitors quality. The orchestrator dispatches at decision points, then gets out of the way.\n- **Do NOT override individual skill guardrails.** If `simplify-and-harden` has a 20% budget cap, the orchestrator respects it.\n- **Do NOT force skills on trivial tasks.** The pipeline exists to help, not to slow down simple work.\n\n## Pipeline Variant Details\n\nFor complete step-by-step walkthroughs of each variant including hybrid scenarios and session resume: read `references\u002Fpipeline-variants.md`.\n","# Skill Pipeline\n\nThe conductor, not a player. This skill classifies tasks, selects the pipeline variant, calibrates depth, and orchestrates handoffs between skills. It produces no artifacts of its own — its output is routing decisions that activate other skills.\n\n## Task Classification\n\nOn every coding task, classify before acting. Evaluate scope signals and map to a task class.\n\n**Input signals:** file count, task description, existing plan\u002Fhandoff files, batch indicators, CI environment.\n\n```\nTask received\n  │\n  ├─ Trivial (typo, rename, version bump)\n  │  → No skills. Just do it.\n  │\n  ├─ Small (isolated fix, single-file, \u003C10 logic lines)\n  │  → verify-gate + simplify-and-harden\n  │\n  ├─ Medium (feature in known area, 2-5 files)\n  │  → intent-framed-agent + verify-gate + simplify-and-harden\n  │\n  ├─ Large (complex refactor, new architecture, unfamiliar codebase, high-risk logic)\n  │  → Full standard pipeline\n  │  → Recommend \u002Fplan-interview before starting\n  │\n  ├─ Long-running (multi-session, high context pressure, prior handoff exists)\n  │  → Full standard pipeline with context-surfing as critical skill\n  │\n  └─ Batch (multiple features from spec, 5+ discrete tasks, issue triage)\n     → Team-based pipeline (agent-teams-simplify-and-harden)\n```\n\nWhen uncertain, start with Medium. Add skills if drift or quality issues appear mid-task.\n\nFor detailed heuristics, edge cases, and examples: read `references\u002Fclassification-rules.md`.\n\n## Pipeline Selection\n\nRoute task class to the right variant:\n\n| Task Class | Variant | Rationale |\n|------------|---------|-----------|\n| Trivial | None | No overhead needed |\n| Small | Standard (minimal) | Verify + S&H only |\n| Medium | Standard (partial) | Scope monitoring + verify + review |\n| Large | Standard (full) | Full inner loop with planning |\n| Long-running | Standard (full) | Context-surfing is critical |\n| Batch | Team-based | Breadth over depth |\n| CI environment | CI | Headless review |\n| Periodic | Outer loop | Cross-session improvement |\n\n**Heuristic:** Standard pipeline for **depth** (single complex feature). Team-based pipeline for **breadth** (batch of tasks). CI pipeline when `CI=true` or `GITHUB_ACTIONS=true`.\n\n## Activation Sequences\n\n### Standard Pipeline (Inner Loop)\n\n```\npre-flight-check (SessionStart hook — surfaces prior learnings)\n  → classify\n  → (recommend \u002Fplan-interview if Large or Long-running)\n  → intent-framed-agent (at planning-to-execution transition)\n  → context-surfing (auto-activates when intent frame + plan exist; concurrent with intent monitoring)\n  → [IMPLEMENTATION]\n  → verify-gate (compile + test + lint; fix loop if red)\n  → simplify-and-harden (post-completion, if non-trivial diff)\n  → self-improvement (on errors, corrections, or S&H learning candidates)\n```\n\n**Skill-by-class activation:**\n\n| Skill | Trivial | Small | Medium | Large | Long-running |\n|-------|---------|-------|--------|-------|-------------|\n| pre-flight-check | Hook | Hook | Hook | Hook | Hook |\n| plan-interview | - | - | - | Recommend | Recommend |\n| intent-framed-agent | - | - | Activate | Activate | Activate |\n| context-surfing | - | - | - | Activate | Critical |\n| verify-gate | - | Activate | Activate | Activate | Activate |\n| simplify-and-harden | - | If non-trivial | If non-trivial | If non-trivial | If non-trivial |\n| self-improvement | On error only | On error only | On error\u002Fcompletion | On error\u002Fcompletion | On error\u002Fcompletion |\n\n### Team-Based Pipeline\n\n```\nclassify (Batch)\n  → (recommend \u002Fplan-interview if no spec exists)\n  → agent-teams-simplify-and-harden\n    ├─ Team lead emits Intent Frame #1\n    ├─ Phase 1: parallel implementation agents\n    ├─ verify-gate (compile + test + lint)\n    ├─ Phase 2: parallel audit agents (simplify, harden, spec)\n    ├─ Fix loop (up to 3 audit rounds)\n    └─ Learning loop output\n  → self-improvement\n```\n\n### CI Pipeline\n\n```\nclassify (CI detected)\n  → simplify-and-harden-ci (headless scan, PR changed files only)\n  → self-improvement-ci (pattern aggregation, promotion recommendations)\n```\n\n### Outer Loop Pipeline\n\nThe outer loop runs across sessions, not within them. Trigger on cadence (weekly, sprint boundary) or when `pre-flight-check` surfaces promotion-ready patterns.\n\n```\nlearning-aggregator (read .learnings\u002F, find patterns, rank promotion candidates)\n  → harness-updater agent (apply promotions to CLAUDE.md, AGENTS.md, copilot-instructions.md)\n  → eval-creator (create permanent test cases from promoted patterns)\n  → eval-creator run (regression check on all existing evals)\n```\n\n**When to trigger the outer loop:**\n- Weekly: recommended minimum cadence\n- Sprint boundary: after a burst of sessions\n- When `pre-flight-check` reports promotion-ready count > 3\n- After a significant incident or recurring failure\n- Manually: user invokes `\u002Flearning-aggregator`\n\n**Outer loop is always human-gated.** `learning-aggregator` produces a gap report. `harness-updater` shows diffs for approval. No automatic writes to instruction files without human review.\n\n## Depth Calibration\n\nNot just which skills — how deep each goes:\n\n| Dimension | Small | Medium | Large | Long-running | Batch |\n|-----------|-------|--------|-------|-------------|-------|\n| Pre-flight check | Hook | Hook | Hook | Hook | Hook |\n| Planning passes | 0 | 0-1 | 1-2 | Deep iterative | Per-task or umbrella |\n| Intent frame | - | Single frame | Full frame + monitoring | Full + handoff | Team lead frame |\n| Context-surfing | - | - | Active | Critical (exit protocol ready) | Lightweight drift checks |\n| Verify-gate | Compile + test | Compile + test | Compile + test + lint | Compile + test + lint | Compile + test (per round) |\n| S&H budget | 20% diff, 60s | 20% diff, 60s | 20% diff, 60s | 20% diff, 60s | 30% team growth cap |\n| Audit rounds (teams) | - | - | - | - | Up to 3 |\n| Self-improvement | Error-triggered | Error-triggered | Error + S&H feed | Error + S&H feed | Error + teams feed |\n\n## Handoff Rules\n\nArtifacts flow between skills. The orchestrator ensures each skill receives what it needs.\n\n**Key handoffs:**\n\n1. **Plan file** (`docs\u002Fplans\u002Fplan-NNN-\u003Cslug>.md`) — produced by `plan-interview`, consumed by `intent-framed-agent` (context), `context-surfing` (wave anchor), `agent-teams` (task extraction).\n\n2. **Intent Frame** — produced by `intent-framed-agent`, consumed by `context-surfing` (wave anchor strengthening). Copied into handoff files on drift exit.\n\n3. **Handoff file** (`.context-surfing\u002Fhandoff-[slug]-[timestamp].md`) — produced by `context-surfing` on drift exit, consumed by next session for resume.\n\n4. **Verify-gate signal** — produced by `verify-gate` (pass\u002Ffail + diagnostics), consumed by `simplify-and-harden` (only activates after green gate) and fix loop (on failure).\n\n5. **Learning candidates** (`learning_loop.candidates`) — produced by `simplify-and-harden` and `agent-teams`, consumed by `self-improvement` for pattern tracking.\n\n6. **Learning entries** (`.learnings\u002F*.md`) — produced by `self-improvement`, consumed by `learning-aggregator` for cross-session analysis and by `pre-flight-check` at session start.\n\n7. **Gap report** — produced by `learning-aggregator`, consumed by `harness-updater` agent for promotion and `eval-creator` for test case generation.\n\n8. **Eval cases** (`.evals\u002Fcases\u002F*.md`) — produced by `eval-creator`, consumed by regression runs and surfaced by `pre-flight-check`.\n\n**Precedence:** If `context-surfing` and `intent-framed-agent` both fire simultaneously, context-surfing's exit takes precedence. Degraded context makes scope checks unreliable.\n\nFor the full artifact\u002Fsignal\u002Fbudget table: read `references\u002Fhandoff-matrix.md`.\n\n## Decision Points\n\nThe orchestrator intervenes at these moments:\n\n### Task Arrival\nClassify the task. Select pipeline variant and depth. Emit routing decision. If Large\u002FLong-running, recommend `\u002Fplan-interview`. If Batch, recommend team-based variant.\n\n### Plan Approval\nWhen user approves a plan from `plan-interview`, flow directly into the execution stage — no separate \"should I proceed?\" prompt. This means activating `intent-framed-agent` to emit an Intent Frame. The intent frame itself still requires user confirmation before coding begins (that confirmation is part of `intent-framed-agent`, not an extra gate). Populate task tracking with checklist items.\n\n### Planning-to-Execution Transition\nWhen no plan-interview was used and the user signals readiness (\"go ahead\", \"implement this\", \"let's start\"), activate `intent-framed-agent`. Emit Intent Frame. Wait for user confirmation of the frame before coding.\n\n### Implementation Complete\nActivate `verify-gate` to run compile, test, and lint checks. If any fail, enter the fix loop (up to 3 attempts per phase). Once all checks pass and the diff meets the non-trivial threshold (see `references\u002Fclassification-rules.md`), activate `simplify-and-harden`. If the diff is trivial, signal completion directly after verify-gate passes.\n\n### Drift Detected\nIf `context-surfing` fires a drift exit, stop execution. Write handoff file. If the task was classified below Large, consider re-classifying upward for the next session.\n\n### Session Resume\nCheck for handoff files in `.context-surfing\u002F`. If found, read completely. Re-establish context from handoff. Re-classify if needed. Resume from recommended re-entry point.\n\n## Overrides\n\nUsers can override any routing decision:\n\n- **Force depth:** `depth=small` \u002F `depth=large` — override classification\n- **Force variant:** `variant=teams` \u002F `variant=standard` — override pipeline selection\n- **Skip review:** `--no-review` — skip `simplify-and-harden`\n- **Force planning:** invoke `\u002Fplan-interview` on any task regardless of classification\n- **Skip all skills:** user says \"just do it\" on a non-trivial task — respect the override\n\n## Re-classification\n\nTasks can change class mid-execution. Watch for:\n\n- **Escalation signals:** scope expanded beyond original estimate, many more files affected than expected, `context-surfing` drift exit, `intent-framed-agent` detects significant scope change\n- **De-escalation signals:** task turns out simpler than planned, plan reveals minimal changes needed\n\nWhen re-classification is warranted:\n1. Note the signal that triggered re-classification\n2. Adjust active skills (add or remove pipeline stages)\n3. If escalating to Large: recommend `\u002Fplan-interview` if no plan exists\n4. If de-escalating: drop unnecessary stages, proceed with lighter pipeline\n\n## Anti-Patterns\n\n- **Do NOT re-implement skill logic.** This skill classifies and dispatches. Each individual skill owns its own procedure.\n- **Do NOT auto-invoke plan-interview.** It is a human gate. Recommend it; let the user decide.\n- **Do NOT create a third monitoring layer.** During execution, `intent-framed-agent` monitors scope and `context-surfing` monitors quality. The orchestrator dispatches at decision points, then gets out of the way.\n- **Do NOT override individual skill guardrails.** If `simplify-and-harden` has a 20% budget cap, the orchestrator respects it.\n- **Do NOT force skills on trivial tasks.** The pipeline exists to help, not to slow down simple work.\n\n## Pipeline Variant Details\n\nFor complete step-by-step walkthroughs of each variant including hybrid scenarios and session resume: read `references\u002Fpipeline-variants.md`.\n",{"name":5,"description":7},"main","87afe6ce96220accf50a345dea46784c8c9f210e","YYC3-Skills-技能包\u002FYYC3-Skills-技能包-pskoett-ai-skills\u002Fskills\u002Fskill-pipeline",0,"https:\u002F\u002Fgithub.com\u002FYYC-Cube\u002FYYC3-Knowledge-Base","https:\u002F\u002Fkb.yyc3.top\u002F","https:\u002F\u002Fgithub.com\u002FYYC-Cube","万象归元于云枢 | 深栈智启新纪元","2026-11-10 10:37:32",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":45},[35,38,41],{"tool":36,"command":37},"npx","npx skills add YYC-Cube\u002FYYC3-Knowledge-Base --skill skill-pipeline",{"tool":39,"command":40},"bunx","bunx skills add YYC-Cube\u002FYYC3-Knowledge-Base --skill skill-pipeline",{"tool":42,"command":43},"pnpm","pnpm dlx skills add YYC-Cube\u002FYYC3-Knowledge-Base --skill skill-pipeline","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FYYC-Cube\u002FYYC3-Knowledge-Base\u002Fblob\u002Fmain\u002FYYC3-Skills-技能包\u002FYYC3-Skills-技能包-pskoett-ai-skills\u002Fskills\u002Fskill-pipeline\u002FSKILL.md",true,[47,53,57],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},"references\u002Fclassification-rules.md","REFERENCE","参考文档",4274,"https:\u002F\u002Fgithub.com\u002FYYC-Cube\u002FYYC3-Knowledge-Base\u002Fblob\u002Fmain\u002FYYC3-Skills-技能包\u002FYYC3-Skills-技能包-pskoett-ai-skills\u002Fskills\u002Fskill-pipeline\u002Freferences\u002Fclassification-rules.md",{"path":54,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":55,"sourceHtmlUrl":56},"references\u002Fhandoff-matrix.md",4316,"https:\u002F\u002Fgithub.com\u002FYYC-Cube\u002FYYC3-Knowledge-Base\u002Fblob\u002Fmain\u002FYYC3-Skills-技能包\u002FYYC3-Skills-技能包-pskoett-ai-skills\u002Fskills\u002Fskill-pipeline\u002Freferences\u002Fhandoff-matrix.md",{"path":58,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":59,"sourceHtmlUrl":60},"references\u002Fpipeline-variants.md",6556,"https:\u002F\u002Fgithub.com\u002FYYC-Cube\u002FYYC3-Knowledge-Base\u002Fblob\u002Fmain\u002FYYC3-Skills-技能包\u002FYYC3-Skills-技能包-pskoett-ai-skills\u002Fskills\u002Fskill-pipeline\u002Freferences\u002Fpipeline-variants.md",[62,69,76,83,90,97],{"skillNo":63,"name":64,"nameZh":6,"description":65,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":66,"categories":67,"keywords":68,"publishedAt":19},"01M7CTDWHERCXPQCP0K03A86AS","vercel-sandbox","Run agent-browser + Chrome inside Vercel Sandbox microVMs for browser automation from any Vercel-deployed app. Use when the user needs browser automation in a Vercel app (Next.js, SvelteKit, Nuxt, Remix, Astro, etc.), wants to run headless Chrome without binary size limits, needs persistent browser sessions across commands, or wants ephemeral isolated browser environments. Triggers include \"Vercel Sandbox browser\", \"microVM Chrome\", \"agent-browser in sandbox\", \"browser automation on Vercel\", or any task requiring Chrome in a Vercel Sandbox.","https:\u002F\u002Fgithub.com\u002FYYC-Cube\u002FYYC3-Knowledge-Base\u002Fblob\u002Fmain\u002FYYC3-WebSpec-Web标准\u002FYYC3-WebSpec-Web标准-standards\u002Fbrowser-agent\u002Fskills\u002Fvercel-sandbox\u002FSKILL.md",[],[],{"skillNo":70,"name":71,"nameZh":6,"description":72,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":73,"categories":74,"keywords":75,"publishedAt":19},"01M7CTD8D2BNJY54S65M47HEJA","slack","Interact with Slack workspaces using browser automation. Use when the user needs to check unread channels, navigate Slack, send messages, extract data, find information, search conversations, or automate any Slack task. Triggers include \"check my Slack\", \"what channels have unreads\", \"send a message to\", \"search Slack for\", \"extract from Slack\", \"find who said\", or any task requiring programmatic Slack interaction.","https:\u002F\u002Fgithub.com\u002FYYC-Cube\u002FYYC3-Knowledge-Base\u002Fblob\u002Fmain\u002FYYC3-WebSpec-Web标准\u002FYYC3-WebSpec-Web标准-standards\u002Fbrowser-agent\u002Fskills\u002Fslack\u002FSKILL.md",[],[],{"skillNo":77,"name":78,"nameZh":6,"description":79,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"publishedAt":19},"01M7CTCSSB8ZP3Q58C3YZDZK1W","electron","Automate Electron desktop apps (VS Code, Slack, Discord, Figma, Notion, Spotify, etc.) using agent-browser via Chrome DevTools Protocol. Use when the user needs to interact with an Electron app, automate a desktop app, connect to a running app, control a native app, or test an Electron application. Triggers include \"automate Slack app\", \"control VS Code\", \"interact with Discord app\", \"test this Electron app\", \"connect to desktop app\", or any task requiring automation of a native Electron application.","https:\u002F\u002Fgithub.com\u002FYYC-Cube\u002FYYC3-Knowledge-Base\u002Fblob\u002Fmain\u002FYYC3-WebSpec-Web标准\u002FYYC3-WebSpec-Web标准-standards\u002Fbrowser-agent\u002Fskills\u002Felectron\u002FSKILL.md",[],[],{"skillNo":84,"name":85,"nameZh":6,"description":86,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":89,"publishedAt":19},"01M7CTCBJY76688V4DCZGWFHGQ","dogfood","Systematically explore and test a web application to find bugs, UX issues, and other problems. Use when asked to \"dogfood\", \"QA\", \"exploratory test\", \"find issues\", \"bug hunt\", \"test this app\u002Fsite\u002Fplatform\", or review the quality of a web application. Produces a structured report with full reproduction evidence -- step-by-step screenshots, repro videos, and detailed repro steps for every issue -- so findings can be handed directly to the responsible teams.","https:\u002F\u002Fgithub.com\u002FYYC-Cube\u002FYYC3-Knowledge-Base\u002Fblob\u002Fmain\u002FYYC3-WebSpec-Web标准\u002FYYC3-WebSpec-Web标准-standards\u002Fbrowser-agent\u002Fskills\u002Fdogfood\u002FSKILL.md",[],[],{"skillNo":91,"name":92,"nameZh":6,"description":93,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":94,"categories":95,"keywords":96,"publishedAt":19},"01M7CTC4H0FFGVEAQR9PA4GA1A","agent-browser","Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to \"open a website\", \"fill out a form\", \"click a button\", \"take a screenshot\", \"scrape data from a page\", \"test this web app\", \"login to a site\", \"automate browser actions\", or any task requiring programmatic web interaction.","https:\u002F\u002Fgithub.com\u002FYYC-Cube\u002FYYC3-Knowledge-Base\u002Fblob\u002Fmain\u002FYYC3-WebSpec-Web标准\u002FYYC3-WebSpec-Web标准-standards\u002Fbrowser-agent\u002Fskills\u002Fagent-browser\u002FSKILL.md",[],[],{"skillNo":98,"name":99,"nameZh":6,"description":100,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":101,"categories":102,"keywords":103,"publishedAt":19},"01M7CTBKVFYBG78FDSABSDJTTQ","shadcn","Manages shadcn components and projects — adding, searching, fixing, debugging, styling, and composing UI, including chat interfaces. Provides project context, component docs, and usage examples. Applies when working with shadcn\u002Fui, component registries, presets, --preset codes, or any project with a components.json file. Also triggers for \"shadcn init\", \"create an app with --preset\", \"switch to --preset\", and for building a design system or applying a DESIGN.md or brand spec to shadcn\u002Fui.","https:\u002F\u002Fgithub.com\u002FYYC-Cube\u002FYYC3-Knowledge-Base\u002Fblob\u002Fmain\u002FYYC3-UI-界面组件\u002FYYC3-UI-界面组件-ui\u002Fskills\u002Fshadcn\u002FSKILL.md",[],[]]