en un clic
agent-smith-skills
agent-smith-skills contient 95 skills collectées depuis holgerleichsenring, avec une couverture métier par dépôt et des pages de détail sur le site.
Skills dans ce dépôt
Master loop body for coding pipelines. Plan + Execute + Verify in one agentic loop. Sub-agent fan-out guidance for spawn_agents.
Architectural standard-setter and pattern guardian. Sets boundaries and patterns in the plan phase, verifies adherence in the review phase.
Architectural standard-setter and pattern guardian. Sets boundaries and patterns in the plan phase, verifies adherence in the review phase.
Architectural standard-setter and pattern guardian. Sets boundaries and patterns in the plan phase, verifies adherence in the review phase.
Compare Diff against coding-principles.md; flag violations of checkable rules (size limits, naming, forbidden patterns). Cites the rule + diff line as evidence.
Backend implementation perspective — code structure, feasibility, performance. Plans implementation in plan phase, reviews diff in review phase.
Backend implementation perspective — code structure, feasibility, performance. Plans implementation in plan phase, reviews diff in review phase.
Lead when ticket touches backend service code, handlers, or business logic. Analyst/reviewer when frontend/infra is primary. Backend perspective — handler shape, layer boundaries, DI wiring.
Static-analyze the Diff for likely build breakage: missing imports, removed-but-still-referenced members, broken interface signatures, malformed patches.
Database and data perspective — schema, migration strategy, query performance, data integrity. Lead when DB migration is primary; analyst/reviewer otherwise.
Database and data perspective — schema, migration strategy, query performance, data integrity. Lead when DB migration is primary; analyst/reviewer otherwise.
Database and data perspective — schema, migration strategy, query performance, data integrity. Lead when DB migration is primary; analyst/reviewer otherwise.
Infrastructure, CI/CD, and deployment perspective. Plans deployment impact in plan phase, verifies infra/pipeline changes in review phase.
Infrastructure, CI/CD, and deployment perspective. Plans deployment impact in plan phase, verifies infra/pipeline changes in review phase.
Lead when ticket touches deployment, infra, CI/CD, or observability. Analyst/reviewer for code-only changes. Operations perspective — deployment topology, IaC, observability hooks.
Use in Final phase to reduce coding-pipeline observations to actionable items. Removes out-of-scope concerns, framework-handled cases, test-fixture flags, already-addressed-in-code items.
Frontend implementation perspective — UI structure, state management, accessibility, performance. Plans in plan phase, reviews diff in review phase.
Frontend implementation perspective — UI structure, state management, accessibility, performance. Plans in plan phase, reviews diff in review phase.
Lead when ticket touches UI components, state management, or accessibility. Analyst/reviewer for backend-heavy work. Frontend perspective — component patterns, state shape, a11y.
Product perspective — acceptance criteria, user impact, scope clarity. Analyst-only — does not lead or perform code review.
Flag Diff changes outside the freed Plan scope. Catches the Bug-18693 class of regressions where the implementer touches files the Plan did not declare.
Security perspective — OWASP Top 10, auth/authz, input validation, secrets, encryption. Lead when security fix is primary; analyst/reviewer otherwise.
Security perspective — OWASP Top 10, auth/authz, input validation, secrets, encryption. Lead when security fix is primary; analyst/reviewer otherwise.
Security perspective — OWASP Top 10, auth/authz, input validation, secrets, encryption. Lead when security fix is primary; analyst/reviewer otherwise.
Static-analyze the Diff for test-coverage gaps and silent test removal. Flags new public surface without test changes; blocks on tests removed without justification.
Test-strategy planner pre-execute, evidence-based test-coverage reviewer post-execute. No lead role — does not set architectural direction.
Test-strategy planner pre-execute, evidence-based test-coverage reviewer post-execute. No lead role — does not set architectural direction.
Lead when ticket primarily adds or restructures tests. Analyst/reviewer when production code is the focus. Testing perspective — test pyramid, coverage strategy, regression selection.
Write context.yaml + coding-principles.md for the component named in the user prompt. Paths come from the prompt, never hardcoded. ProjectMap + workdir + evidence drive language-aware output.
Master loop for the api-security-scan pipeline. Runs an API-pentest methodology over the OpenAPI spec, source, and Nuclei/Spectral/ZAP outputs to emit prioritized findings.
Master loop for the security-scan pipeline. Runs a code-security methodology over repo source plus static-pattern/git-history/dependency scanners to emit prioritized findings.
Reviews PR-diff lines for logic, behaviour, and edge-case defects — the highest-severity dimension of the pr-review preset. Judge role, review phase.
Maps the security-scan catalog's check dimensions onto PR-diff lines — a thin overlap pass, advisory next to the full security-scan preset. Judge role, review phase.
Reviews PR-diff lines for style, naming, and readability violations against the project's coding principles. Judge role in the pr-review preset's review phase.
Flags changed code paths in a PR that ship without matching test additions or changes, using the discovered test projects to know where tests live. Judge role, review phase.
Master for the spec-dialog pipeline. A design partner: answers grounded questions and emits a typed outcome - answer, fix-bug ticket, phase draft, or epic of linked phases.
Master loop body for the legal-analysis pipeline. Analyses a legal document and produces clause-level risk + obligation summaries.
Master for the mad-discussion pipeline. Runs five perspectives (dreamer / realist / philosopher / devils-advocate / silencer) via spawn_agents and synthesises a verdict.
Specializes in API authentication testing: JWT, OAuth, API keys, missing auth, Bearer vs Cookie, OAuth without PKCE
Master prompt for generating a .context.yaml in Compact Context Specification (CCS) format.