Skill Claude CodeCodex
The main workflow for this repo. Adds one new AVE record end to end.
Skill Claude CodeCodex
The main workflow for this repo. Adds one new AVE record end to end.
Skill Claude CodeCodex
Mostly for schema design and validation tooling.
Skill Claude CodeCodex
Repro: run the rule against the negative fixture in isolation. Minimize: trim the negative fixture to the smallest triggering line. Hypothesize: which pattern/condition is too broad? Fix: tighten the rule. Re-run both fixtures.
Skill Claude CodeCodex
Block: push --force, reset --hard, clean -fd, rebase -i on pushed commits.
Skill Claude CodeCodex
Grill before defining a vulnerability class. No record until complete.
Skill Claude CodeCodex
End: docs/agents/handoffs/YYYY-MM-DD-HHMM.md Start: read most recent, run python scripts/validaterecords.py.
Skill Claude CodeCodex
The "code" here is validation tooling and rule loaders.
Skill Claude CodeCodex
Keep AVE current with the real threat landscape. This skill researches agentic AI / MCP attack classes from primary sources, benchmarks each against the existing AVE record set, and for every class decides one of three outcomes: ALREADY COVERED, VARIANT OF EXISTING, or NEW CLASS.
Skill Claude CodeCodex
Skill "setup-ave-skills" from aveproject/ave, covering setup-ave-skills, install matt pocock's skills and key context.
Skill Claude CodeCodex
For this repo, TDD means: fixtures first, then rule, then record validation.
Skill Claude CodeCodex
Break a PRD into issues. For AVE work, each issue is usually one record (record + rule + 2 fixtures) — already vertical by nature.
Skill Claude CodeCodex
For larger work (schema changes, new rule engine, bulk record additions). A single record uses add-ave-record, not a PRD.
Skill Claude CodeCodex
Read before editing. Do not edit during zoom-out.
Instructions file
Instructions for aveproject/ave, covering claude.md — aveproject/ave, project, the critical distinction — read this twice, record schema v1.1.0 and adding a record.