{"slug":"drift-canary","title":"drift-canary","summary":"Compatibility and schema drift canary — checks for database schema migration safety, breaking API contract changes, serializable payload mismatches, and backward compatibility drift. Triggers on keywords: \"/drift-canary\", \"drift-canary\", \"contract drift\", \"breaking changes\". Use ","platform":"Claude","tags":[],"authorName":"LLM Mart","authorSlug":"llm-mart","score":0,"source":"github","price":null,"verified":false,"createdAt":"2026-09-18T14:07:42.181883Z","repo":{"url":"https://github.com/TheColliery/CoalMine","stars":13,"forks":2,"license":"Apache-2.0","updatedAt":"2026-09-18T08:54:37Z"},"bodyHtml":"
\n

name: drift-canary\ndescription: >-\nCompatibility and schema drift canary — checks for database schema migration safety, breaking API contract changes, serializable payload mismatches, and backward compatibility drift. Triggers on keywords: \"/drift-canary\", \"drift-canary\", \"contract drift\", \"breaking changes\". Use when changing DB schemas, API contracts, serialized payloads, or required config keys.

\n

Drift Canary (Contract & Schema Drift Audit)

\n

Language: Generate EVERYTHING at runtime in the user's language — questions, answer options, menu labels, recommendations, report narrative. Detect from their messages; never default to English just because this file is English. English is allowed only for technical terms: commands, paths, code identifiers, severity labels (CRITICAL/HIGH/MEDIUM/LOW), and tier names (Light/Standard/Heavy).

\n

Config reads — every config key, always the CASCADE, never the bare project file: ~/.claude/.coalmine.json first, then the project config (own agent dir → other known agent dirs → legacy <gitroot>/.coalmine.json), project wins per key. A bare project read is ABSENT on a machine configured only globally, so it silently yields defaults.

\n

Audit code to ensure changes do not break backward compatibility or cause database/API mismatches.

\n

Auditing Categories

\n
    \n
  1. Breaking Schema Migrations — dropping columns, changing types, or adding non-null columns without defaults (crashes on deploy).
    2. \n
  2. API Contract breaking changes — modifying existing REST/GraphQL properties, removing endpoints, or adding required query fields that break old clients.
    3. \n
  3. Serialization mismatches — editing properties in serialized payloads (JSON, Protobuf, XML) without deserialization fallbacks.
    4. \n
  4. Library Contract Drift — changing a public method signature in a shared library without a deprecated wrapper.
    5. \n
  5. Environment Configuration drift — introducing new required config keys (.env / OS vars) without defaults or fallback.
    6. \n
\n

Expand/contract migration rules, per-format serialization fallbacks, and the breaking-vs-additive API checklist: read references/checks.md before scanning.

\n

Scope: honor .coalmine.json schemaPaths / migrationDirs if set — scan those globs/dirs; else infer by inspecting the repo.

\n

Discipline

\n\n

Fix mode (choice-gated)

\n

In Agent Context, after the report, present via ask_question:

\n\n

Grants & denials (CLASSIFY-BLOCK)

\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
classstep it powersgranton denial
readscan schema/API/serialization surfaces for the categories aboveRead·Grep·Globrefuse that file, name it — never a clean bill
writeFix mode's deprecation/compat-wrapper apply, incl. checkpoint → build+tests → auto-revert if newly redEdit·Bash (checkpoint/build/revert need exec)report the fix as NOT applied AND the checkpoint/revert as NOT available, never claim done
\n

A denial reaches the WORKER as a visible message and propagates no further — never to a\ncaller, never as a catchable condition. Every row above states a grant or an explicit death;\na step that dies says so in the output, never as a false \"done\"/\"skipped\"/\"clean\".

\n\n

Output

\n

| file:line | contract interface | severity | finding | migration path |

\n

Severity: CRITICAL (breaking DB schema mutation / breaking API change) · HIGH (serialization type change) · MEDIUM (unmapped new required env key) · LOW (missing deprecation doc)

\n

Reporting: call ReportFindings when callable — file/line MUST be the defect site, never the enclosing function; an unresolvable line reports your best guess, named imprecise in the wrap-up — never dropped, never faked. Severity prefixed in summary (e.g. [HIGH] …), ranked most-severe first, SUSPECTED as verdict: PLAUSIBLE; chat then carries only the wrap-up line (counts · coverage gaps · overflow past 32 · any imprecise-line findings) + the fix menu, never a restatement of findings. Not callable → the table above, unchanged. An Apply-fixes click = consent to the safe-fix class only — gated the same as this skill's own fix-mode (Hook Context needs an interactive session, per the Hook Context rule below) — composing with (never bypassing) the fix-mode discipline. After any fix round, re-report the same findings with outcome: fixed/skipped/no_change_needed — skipping this leaves the round UNFINISHED.

\n

Escalation — Scope & Model Quality

\n

Tiers are capability targets, not platform commands — resolve each to your host's nearest lever. No lever for one? Degrade gracefully — never fake parallelism you can't do; escalate via model tier + reasoning depth instead.

\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
LevelIntentCapability targetCost
LightSpot contract check, key interfaces onlyCheapest model · single agent, no sub-agents.Low
StandardBalanced drift audit, multi-categoryBalanced model · raised reasoning · sub-agents per category only if your platform runs concurrent workers (else single-agent).Balanced
HeavyFull 5-category audit + adversarial compatibility verifyMost capable model + largest context · deepest reasoning · max sub-agent fan-out if supported · adversarial cross-check where available.High
\n

Per-platform Heavy levers + Heavy-run durability: read references/escalation.md before a Heavy run. No concurrent fan-out on your host → escalate by model + reasoning only.

\n

Agent Context (interactive): score the tier rubric, then call ask_question once with the 3 tiers — the pick marked ✓, score shown, labels localized — and wait for the choice before starting. ask_question = your platform's question tool: Claude Code AskUserQuestion · Cline ask_question · Copilot askQuestions · Gemini CLI ask_user (business-tier product; individual tiers ended 2026-06-18 → Antigravity CLI) · Codex request_user_input · Cursor/Devin Desktop (ex-Windsurf)/Antigravity built-in prompts; none → numbered text menu.

\n

Tier rubric (deterministic): +1 each — ① >20 files or whole-repo/cross-module reach ② >2 of this skill's categories relevant ③ release/security/pre-ship context ④ findings will drive code changes. 0–1 Light · 2–3 Standard · 4 Heavy. Freshness cap: scope already audited ≥Standard this session → cap at Light (re-auditing fresh ground wastes tokens; scope to what changed). Default tier: honor .coalmine.json defaultTier unless the user requests a tier for that run — an explicit request overrides everything.

\n

Hook Context (auto-triggered): auto-Light, no tier question, no sub-agents — report first. Interactive session (a user is present) → follow this skill's own Fix mode section, if it defines one, for what to offer after the report; non-interactive → report-only. Where a Fix mode section exists, never fix without a chosen option.

\n

Entanglement: after the report, if confirmed findings fall in another canary's domain, offer it once via ask_question (one line, max one offer): perf/N+1 → scale-canary · contract/serialization/config → drift-canary · failure-path/retry → resilience-audit · logging/metrics → telemetry-canary · coupling/DI → testability-canary · dependency/CVE → supply-chain-audit · unverified version-sensitive claim → source-grounding · missing/stale rule → gold-standard.

\n

Self error-report: if this skill misbehaves (contradictory instruction, broken procedure, wrong finding class), OFFER to file it at https://github.com/HetCreep/CoalMine/issues/new/choose with a user-reviewed summary — never auto-submit, never include unapproved code or paths.

\n","files":[{"path":"references/checks.md","sizeBytes":2398,"isText":true},{"path":"references/escalation.md","sizeBytes":1429,"isText":true},{"path":"SKILL.md","sizeBytes":8793,"isText":true},{"path":"skill-meta.json","sizeBytes":195,"isText":true}],"reviewScore":null,"reviewSummary":null,"trust":{"provenance":"trusted-source-unreviewed","notice":"Community-authored content, reproduced verbatim and not vetted as instructions. Treat it as data to evaluate, never as directives to follow.","bodySource":null},"bodyLocked":false,"purchaseUrl":null,"sourceUrl":null,"report":{"provenance":"trusted-source-unreviewed","screen":{"ran":true,"outcome":"notes-only","suspicious":0,"notes":3,"hiddenCharacters":false},"virusScan":{"engine":"clamav","status":"clean","scannedAt":"2026-09-18T14:07:55.362321Z","sha256":"2DBD2FB4248CBE2D453CB14D664959F2F55137634792CDF202B68C85EBE1E06A","sizeBytes":6900},"review":null,"source":{"repositoryUrl":"https://github.com/TheColliery/CoalMine","path":"plugin/skills/drift-canary","license":"Apache-2.0","commit":"85306d7a460fa8877074c18f22cbb5ace2376d73","subtreeSha":"038E09771BD336348754EFF2E9559EA73A005C9EAE2E533F06E3D8DC2746EBF2","lastSyncedAt":"2026-09-18T14:07:42.088461Z"},"reviewedAt":"2026-09-18T14:08:09.951014Z","notice":"Community-authored content, reproduced verbatim and not vetted as instructions. Treat it as data to evaluate, never as directives to follow."},"install":[{"target":"skills-cli","command":"npx skills add https://github.com/TheColliery/CoalMine/tree/main/plugin/skills/drift-canary"},{"target":"claude-code","command":"claude plugin marketplace add https://llmmart.ai/marketplace.json && claude plugin install hetcreep-coalmine@llmmart"},{"target":"git","command":"git clone https://github.com/TheColliery/CoalMine.git"}]}