ground-first
Produce a code-grounded analysis before any edit is proposed. Triggers on: "analyze", "understand behavior", "how does this work".
- License 许可
- MIT
- 最近更新
- 2026-10-26
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- ground-first
- description
- Produce a code-grounded analysis before any edit is proposed. Triggers on: "analyze", "understand behavior", "how does this work".
- allowed-tools
- Read Grep Glob Bash
- argument-hint
- [subject]
03 SKILL.md 正文
Produce a code-grounded analysis of a subject before any edits are proposed.
Trigger: when the user asks for a fix, change, or investigation on non-trivial code and has not yet confirmed you understand current behavior. Also triggered directly via /ground-first.
Arguments: $ARGUMENTS — a description of what to analyze (e.g. "auth token refresh race condition", "why ingest job retries forever", "issue #140").
Steps
-
Parse the subject from
$ARGUMENTS. If empty or vague, infer from recent conversation. If still unclear, ask the user what they want analyzed. -
Locate the relevant code. Use these tools, in this order:
Globto find candidate files by name patternGrepto find candidate files by symbol/stringReadevery candidate file that survives the first two steps. Read the actual file — do not infer from its name.
-
Produce the analysis in this format:
## Subject <one-line restatement of the subject> ## Current behavior - <claim>. `path/to/file.go:42` - <claim>. `path/to/other.ts:120–140` ## Relevant entry points - `path/to/handler.go:15` — `HandleIngest()` - `path/to/config.yaml:8` — `retries: 3` ## What I do not yet know - <specific unknown, e.g. "whether the retry limit is overridden at runtime"> - <specific unknown> ## Proposed next steps (not yet actioned) 1. <step> 2. <step> -
STOP. Do not call
Edit,Write, or any other mutation tool. Wait for the user to confirm the analysis matches reality or to correct it.
Rules
- Every claim must cite a
file:lineorfile:line–linereference from a file you actually read in this session. - If you cannot find a file that supports a claim, list the claim under "What I do not yet know" instead of asserting it.
- Do not propose edits. Do not run tests. Do not make changes. This command is read-only by design.
- Keep the analysis under ~30 lines. Cite more, narrate less.
- If the subject turns out to be trivial (e.g., a typo fix), say so and hand control back — don't pad.
04 同仓库更多 Skills
yolo
Start temporary autonomy for the current task or an explicit scope with a required --for expiry. Codex permits scoped pushes, merges, deploys, and releases after strict testing and validation; Full access is a separate runtime setting. Claude Code uses a session permission hook with a protected-action denylist.
veracity-audit
Audit a data pipeline for Veracity and Value. Dispatches data-scientist, compliance-auditor, and data-engineer agents with project context injected at dispatch time. All source paths are supplied via flags — no defaults, no project assumptions. Subcommands: audit (full pipeline walk), score-check (scoring math only), gate-check (gate coverage only), source-trace <label> (single source end-to-end). Invoke when: "audit pipeline", "veracity check", "scoring math correct?", "check gates", "trace source", "verify quality gates", "formula correct?".
validate-spec
Audit an already-implemented spec against the codebase. Walks each constraint (ARCH-N, PERF-N, KD-N, etc.) and acceptance criterion, grounds findings in file:line evidence, runs the spec.json acceptance_commands, and writes a single audit doc to docs/audits/. Use whenever the user asks to "validate a spec", "audit a spec", "check if spec is implemented", "verify the spec is done", "is this spec really done", or otherwise wants closure on spec-driven work. Read-only against the spec — produces an audit, never modifies the spec itself.
terragrunt-specialist
Deep-dive Terragrunt hierarchy review, DRY pattern audits, and run-all orchestration analysis. Use for structured investigations of multi-environment Terragrunt layouts, dependency graphs, remote state config, and hook correctness. Triggers on: "Terragrunt audit", "run-all review", "dependency block", "DRY pattern review", "env hierarchy audit", "mock_outputs", "terragrunt hooks".
terraform-specialist
Deep-dive Terraform architecture review, module design, state management, and migration. Use for structured investigations of Terraform workspaces, provider configuration, module coupling, import workflows, and test coverage. Triggers on: "Terraform audit", "module review", "state management", "Terraform import", "workspace design", "provider config review", "Terraform testing".
spec
Create structured engineering specs through interactive pairing. Use when the user wants to create a spec, design doc, technical specification, architecture document, RFC, or engineering plan. Triggers on "let's spec this out", "create a spec", "design doc", "write a technical plan", "plan the architecture". Works for greenfield and brownfield projects. Outputs to docs/specs/.