From workflow
Read-only drift detector that diffs SPEC.md against current code and reports violations grouped by severity. Triggers on 'check drift', 'audit the spec', or 'verify invariants'.
How this skill is triggered — by the user, by Claude, or both
Slash command
/workflow:checkThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Pure diagnostic. Reports violations. Writes nothing. User decides remedy.
Pure diagnostic. Reports violations. Writes nothing. User decides remedy.
SPEC.md. If missing → "no spec, nothing to check." Stop.§V → check invariants only (default)§I → check interfaces§T → audit task status vs code--all → all threeFor each V:
For each I item:
For each T:
x: verify claimed work present.~: note as in-progress..: note as pending.x rows with no evidence as STALE.Caveman. Grouped by severity.
## §V drift
V2 VIOLATE: auth/mw.go:47 uses `<` not `≤`. see §B.1.
V5 UNVERIFIABLE: no test covers ∀ req path.
## §I drift
I.api DRIFT: POST /x returns `{result}` not `{id}`. route.go:112.
I.cmd MISSING: `foo bar` absent from cli/*.go.
## §T drift
T3 STALE: status `x`, no middleware file exists.
## summary
2 violate. 1 missing. 1 stale. 1 unverifiable.
next: spec skill with `bug:` or fix code at cited lines.
End report with one-line hint per class:
bug: <V.n> or fix code.§T.n if task exists; else spec skill amend §T.amend §T to uncheck.amend §I to document, or delete code.Never invoke fixes. Report only.
npx claudepluginhub kushal9889/claude-plugins --plugin workflowDiffs SPEC.md against current code and reports violations grouped by severity. Read-only diagnostic for spec-code alignment.
Detects documentation-code drift via grep/AST fact extraction from CLI commands, APIs, env vars, public functions and LLM-graded semantic analysis of mismatches.
Verifies implementation completion by running tests, code hygiene review, spec compliance validation, and drift checks; blocks claims on failures. Use before commits or merges.