Eric GullicksonandClaude Fable 5
b601dcd5ce
Deploy to Staging / Build Images (pull_request) Successful in 4m33s
Deploy to Staging / Deploy to Staging (pull_request) Successful in 45s
Deploy to Staging / Verify Staging (pull_request) Successful in 3s
Deploy to Staging / Notify Staging Failure (pull_request) Skipped
Deploy to Staging / Notify Staging Ready (pull_request) Successful in 3s
Adds five role agents under .claude/agents/, each pinned to a model in
frontmatter (sonnet for developer/debugger/operator/tech-writer, opus
for quality-reviewer) and directed to read the relevant mvp-* skills
before acting. Replaces the model enforcement lost when c239bb9 deleted
enforce-agent-model.sh. Updates .claude/CLAUDE.md and root CLAUDE.md
indexes.
Co-Authored-By: Claude Fable 5 <[email protected]>
46 lines
4.3 KiB
Markdown
46 lines
4.3 KiB
Markdown
# .claude/
|
|
|
|
## Subdirectories
|
|
|
|
| Directory | What | When to read |
|
|
| --------- | ---- | ------------ |
|
|
| `agents/` | Role subagent definitions with pinned models (see Agents table below) | Delegating work to a subagent; changing a model pin |
|
|
| `skills/` | 16 `mvp-*` reusable knowledge skills, one directory each (see Skills table below) | Loading project-specific knowledge before a task |
|
|
| `output-styles/` | Output formatting templates (`direct.md`) | Customizing response style |
|
|
| `tdd-guard/` | TDD-guard tool state (test/modification tracking data) | Debugging TDD-guard behavior |
|
|
|
|
## Agents (`agents/`)
|
|
|
|
Each agent is `agents/{name}.md`; frontmatter pins its model. Agents are role executors - each one's prompt directs it to read the relevant `skills/` SKILL.md files before acting, so skills stay the single source of domain knowledge.
|
|
|
|
| Agent | Model | Role | Primary skills read |
|
|
| ----- | ----- | ---- | ------------------- |
|
|
| `developer` | sonnet | Implement scoped code changes, self-gate lint/type-check/tests | mvp-architecture-contract, mvp-change-control |
|
|
| `debugger` | sonnet | Root-cause a symptom with evidence; no fixes unless asked | mvp-debugging-playbook, mvp-failure-archaeology |
|
|
| `quality-reviewer` | opus | RULE 0/1/2 merge-gate review; read-only toolset | mvp-change-control, mvp-validation-and-qa |
|
|
| `operator` | sonnet | Deploys, rollbacks, backups, CI workflow changes | mvp-run-and-operate, mvp-deploy-safety-campaign |
|
|
| `tech-writer` | sonnet | Doc-sync and new docs under house style | mvp-docs-and-writing, mvp-launch-readiness |
|
|
|
|
## Skills (`skills/`)
|
|
|
|
Each skill is `skills/{name}/SKILL.md`; its frontmatter `description` states exactly when to load it — this table is a condensed pointer, not a substitute for reading it.
|
|
|
|
| Skill | What | When to read |
|
|
| ----- | ---- | ------------- |
|
|
| `mvp-architecture-contract/` | Load-bearing architecture decisions, invariants, and known weak points | Designing, extending, or reviewing a feature or refactor |
|
|
| `mvp-build-and-env/` | Fresh-checkout setup and build/test/install failure triage | Any local build, install, or test command fails |
|
|
| `mvp-change-control/` | Issue/branch/PR workflow, RULE 0/1/2 quality-review taxonomy, label discipline | Making any repo change, reviewing code, asking "is this safe to merge" |
|
|
| `mvp-config-and-secrets/` | Env vars, config fields, Docker secrets, feature flags, tier gates | Adding/changing config or secrets; startup config errors |
|
|
| `mvp-debugging-playbook/` | Symptom-to-root-cause triage index | Diagnosing a live bug from its symptom |
|
|
| `mvp-deploy-safety-campaign/` | CI/CD safety gaps (no tests/lint/security gate, `:latest` tag hazard) and the plan to close them | Working on CI/CD pipeline safety |
|
|
| `mvp-diagnostics-and-logging/` | Grafana/Loki/LogQL, health checks, redis-cli, shipped diagnostic scripts | Observing or measuring the running system instead of guessing |
|
|
| `mvp-docs-and-writing/` | Doc trust map, house style, doc-sync discipline, temporal-contamination rule | Reading or writing any documentation, including CLAUDE.md files |
|
|
| `mvp-failure-archaeology/` | Chronicle of past incidents, dead ends, and reverts with root cause and evidence | Before re-investigating a familiar-smelling bug or deleting old-looking code |
|
|
| `mvp-launch-readiness/` | Pre-launch blockers and guardrails for external/user-facing claims | Assessing launch readiness or writing a public claim about the product |
|
|
| `mvp-ocr-gemini-pipeline/` | OCR engine stack, Gemini semantic extraction, VIN decode, WIF auth chain | Working on OCR, VIN decode, or email-ingestion features |
|
|
| `mvp-proof-and-analysis-toolkit/` | Prove-it methods for high-stakes decisions and stuck bugs | Before committing to an architectural/design decision, or after a failed fix attempt |
|
|
| `mvp-research-frontier/` | Open problems and methodology for proposing/validating new work | Seeding the next piece of work; the issue tracker is empty |
|
|
| `mvp-run-and-operate/` | Deploy, rollback, blue/green switch, backup/restore procedures | Operating staging or production environments |
|
|
| `mvp-validation-and-qa/` | Evidence bar and definition of done; real test-suite map | Before claiming a change is done, tested, or ready to merge |
|
|
| `mvp-vehicle-domain-reference/` | Vehicle domain theory: VIN validation/decode, MPG/unit conversion, tiers | Working on VIN, fuel-efficiency, unit-conversion, or tier-limit logic |
|