From audit-fleet
Run the audit-fleet validation gate for fixed reports, section contract, and finding field schema.
How this command is triggered — by the user, by Claude, or both
Slash command
/audit-fleet:validate [--mode strict|balanced] [--reports-dir .audit-fleet/reports] [--json-dir .audit-fleet/reports-json] [--bundle .audit-fleet/audit-bundle.json]This command is limited to the following tools:
The summary Claude sees in its command listing — used to decide when to auto-load this command
# Audit Fleet Validate
Validate markdown and JSON artifacts against the audit-fleet contract.
## Fixed Contract Checks
- Required report files are exactly `00` through `13` lane files.
- Required markdown sections in each report (ordered):
- `Executive Summary`
- `Findings`
- `Quick Wins`
- `High-Impact Expansions`
- Every finding object must contain exact keys:
- `finding_id`, `severity`, `dimension`, `evidence`, `impact`, `recommendation`,
`effort`, `owner`, `dependencies`, `confidence`, `acceptance_criteria`
- Allowed values:
- `severity`: `critical|warning|info`
- `...Validate markdown and JSON artifacts against the audit-fleet contract.
00 through 13 lane files.Executive SummaryFindingsQuick WinsHigh-Impact Expansionsfinding_id, severity, dimension, evidence, impact, recommendation,
effort, owner, dependencies, confidence, acceptance_criteriaseverity: critical|warning|infoconfidence: high|medium|loweffort: S|M|Lbalanced (default): warnings are emitted but do not fail the command.strict: any warning or error fails the command..audit-fleet/validation-result.jsonreports-check.py --reports-dir <reports-dir> --mode <mode> --output .audit-fleet/reports-check.jsonschema-validate.py --reports-dir <reports-dir> --json-dir <json-dir> --bundle <bundle> --mode <mode> --output .audit-fleet/validation-result.json/audit-fleet:validate
/audit-fleet:validate --mode strict
/audit-fleet:validate --reports-dir .audit-fleet/reports --json-dir .audit-fleet/reports-json --bundle .audit-fleet/audit-bundle.json
npx claudepluginhub nsalvacao/nsalvacao-claude-code-plugins --plugin audit-fleet/verifyConfirms fixes from a previous audit report by checking each finding against current code. Accepts a report file path or inline report text.
/validateRuns an AI audit of code against SERVICE_MAP.yaml, discovering service contracts and linting each service. Produces a markdown report with BLOCK/WARN/INFO findings. Read-only.
/allValidates tool usage, documentation consistency, cross-references, best practices, and execution flow, producing a score, prioritized findings, and detailed report.
/validateRuns data validation checks on CSV, Parquet, or JSON datasets using optional YAML suite or auto-generated rules, producing Markdown report with summary, failed checks, and quality score.
/validateValidates CLI harness for software at given path or GitHub repo against HARNESS.md standards: directory structure, files, Click CLI, core modules, tests, and docs.
/validateValidates a bug bounty finding through a structured 7-question gate and 4-gate checklist, killing weak submissions before report writing.