Skip to content

haksolot/ank / skill/drift

ank-drift

Verified

Audit the decisions in .ank/ against the current code and report what no longer holds. Use when asked whether ADRs, specs, or tasks are still accurate, after a milestone, or when the corpus and the code seem to disagree.

ank-drift

Decisions age. The corpus says what was true when each one was ratified; the code says what is true now. This skill compares the two and reports. It never rewrites.

The ank skill is the contract and applies here in full. This file adds the audit policy only.

Gather

ank review              what is proposed, and which scopes have gone dead
ank check               what is already known to be wrong
ank find --type adr     the decisions; --type spec, the rules
ank graph               an order that may no longer make sense
ank log <id>            what executors recorded against the entity

Logs are the richest source: a discrepancy entry is a finding somebody already made for you, recorded mid-task and read by nobody since.

Then read the code each scope covers. ank scope names what claims to govern a file; the file says whether the claim still holds.

Judge

For each accepted decision, three questions:

  • Does the constraint still describe the system?
  • Does the assumption it rests on still hold?
  • Does anything now contradict it: another entity, the code, a test?

Evidence or it is not a finding. Name the entity, quote the assumption, point at the code, test or log that breaks it.

Report

Findings ordered by impact, each in this shape:

ADR-xxxx  assumed:    what the decision took as true
          measured:   what the repository shows now
          evidence:   file, test, or log entry
          recommend:  supersede, keep, or needs a decision

Never edit an accepted entity and never write the superseding one uninvited. Superseding is planning: hand the findings to the human, and what they decide enters through ank-plan.

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: 468kUpdated 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: 468kUpdated Claude CodeCodexVerifiedInstall
rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics.Stars: 468kUpdated 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: 194kUpdated Claude CodeCodexVerifiedInstall