Aidd Converge — Security skill for Claude Code
AIDD Step 6 — independent, specialized audit (check_spec.py once before the closing audit).
How to install Aidd Converge
Installs to ~/.claude/skills/quimhen-aidd-aidd-converge/SKILL.md
mkdir -p ~/.claude/skills/quimhen-aidd-aidd-converge && curl -fsSL https://raw.githubusercontent.com/quimhen/aidd/HEAD/commands/aidd-converge.md -o ~/.claude/skills/quimhen-aidd-aidd-converge/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Aidd Converge does
description: AIDD Step 6 — independent, specialized audit (check_spec.py once before the closing audit)
Invoke the `aidd` skill and run **Step 6 (Converge)** for: $ARGUMENTS
- Run
python scripts/check_spec.py specs/[###-feature]/ONCE before the closing audit and after the last spec/plan/tasks edit, not before every pass — let the stdlib gap-checker catch mechanical drift once. - Closing verification follows rule E9: one
aidd verifyper spec at close and `tests/run_parallel.py --ch
Alternatives in Security
- Spec Check — Audit GWT acceptance test specs for implementation leakage 149 ★
- Release Audit — Use right before cutting a release — it spawns ONE fresh sibling session via CCC that audits only the JUDGMENT 130 ★
- UX A11y — WCAG 2.1 AA accessibility audit plus common-courtesy checks beyond the spec 66 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Icode Skill
A 6-step coding workflow for Claude Code — Plan → Review → Finalize → Code → Deep Check → Audit. Run all at on
Aidd Audit
AIDD Step 1/1.5 — mechanical mockup audit + Flowmap (interactive flow + pseudocode)
Fleet Audit
Independent, read-only adversarial drift audit of the repository (or a path) against AGENTS.md and the spec
Code Review Feature
Runs the security and quality reviewer subagents in parallel over a Spendly feature and produces a unified rep
Review Implementation
Independent implementation audit of uncommitted code by Codex CLI against a plan doc, then apply agreed fixes
Claude Next
One-key handoff for deep Claude Code sessions. Independent fresh-context audit + drift check + forced uncertai
Related Agents
Spec Test Author
Writes scenario tests from OpenSpec specs, with restricted context kept independent from the implementer. Disp
Math Reviewer
Independent verifier for mathematical and numerical work. Mode "spec" recomputes and checks a math-spec deriva
Kordal Reviewer
Independent reviewer of one diff on one axis, Standards or Spec, for the review step of the kordal delivery wo