Skip to content

The-Artificer-of-Ciphers-LLC/skills-from-the-artificer / adr-phase-coverage

adr-phase-coverage

Verified

Use when a user needs to turn a design doc, ADR, RFC, or spec into a phased implementation plan — or verify that an existing phase breakdown actually covers everything the design promised. The failure mode this catches: Phase N defers a deliverable to Phase N+1, Phase N+1 never claims it, and it ships as promised-but-not-built. Trigger when users say: 'break this into phases', 'turn this ADR/RFC into a roadmap', 'is my phase plan complete', 'what slips between phases', 'sanity check my phase breakdown', 'audit this epic for coverage gaps', or ask to sequence any architectural work into milestones. Also trigger on `/adr-phase-coverage`. Do NOT trigger for: authoring a new ADR, updating ADR rationale, writing a PRD, debugging a stuck deploy, or dependency analysis.

evals/evals.json

View on GitHub
{
  "skill_name": "adr-phase-coverage",
  "evals": [
    {
      "id": 1,
      "prompt": "I'm about to lock the implementation plan for our self-service API key feature. The ADR is at evals/fixtures/adr-0007-apikeys.md and the draft phase roadmap is at evals/fixtures/ROADMAP-apikeys.md. Before I commit to it, check that the phase breakdown actually delivers everything the ADR promises — I don't want anything falling through the cracks between phases.",
      "expected_output": "Builds a decision->phase traceability matrix and flags that the user-facing 'create/rotate/revoke from the dashboard' capability is owned/wired by no phase: Phase 2 defers the dashboard UI to Phase 3, but Phase 3 only owns rate limiting (D5) and its acceptance criteria never mention the dashboard. Reports orphan-decision + dangling-deferral + keystone-unwired (the dashboard is a user-facing surface with no phase that wires it). Recommends giving an owning phase an explicit 'user can create/rotate/revoke keys from the dashboard' acceptance criterion before locking.",
      "files": ["evals/fixtures/adr-0007-apikeys.md", "evals/fixtures/ROADMAP-apikeys.md"]
    },
    {
      "id": 2,
      "prompt": "/adr-phase-coverage — we wrote an RFC for an offline-sync layer (local write queue, conflict resolution, a background sync worker, and a 'Sync now' button + a sync-status indicator in the app). I've sketched 4 phases: (1) local write queue, (2) conflict resolution engine, (3) background sync worker, (4) telemetry + metrics. Turn the RFC into a sound phased plan and make sure the user-facing pieces actually get wired, not just the engine.",
      "expected_output": "Notices the RFC promises two user-facing surfaces (the 'Sync now' button and the sync-status indicator) that none of the 4 proposed phases own or wire — phases 1-3 are all engine/infra layers and phase 4 is telemetry. Flags this as a keystone-unwired/orphan risk and recommends an explicit phase (or acceptance criteria) that wires the button + indicator to the sync worker, ideally via a thin end-to-end slice early rather than horizontal layers.",
      "files": []
    },
    {
      "id": 3,
      "prompt": "Can you audit this migration plan? We're moving auth from session cookies to JWT across 5 phases: (1) issue JWTs alongside cookies, (2) validate JWTs in middleware, (3) migrate the mobile client, (4) migrate the web client, (5) remove cookie code. The original design doc also said admins must be able to force-revoke a user's active tokens. Did we miss anything?",
      "expected_output": "Catches that the design's 'admins can force-revoke active tokens' requirement is not owned by any of the 5 phases (a JWT revocation list / admin action is neither in the issue/validate phases nor the client migrations nor the cleanup). Reports it as an orphaned decision and notes the admin force-revoke is a user-facing capability needing an owning phase that wires the admin surface + the revocation check in middleware.",
      "files": []
    }
  ]
}
rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first.Stars: 196kUpdated Claude CodeCodexVerifiedInstall