kb-markdown-accessibility
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [SAFE]: The skill is configured as a reference-tier module and explicitly disables direct model invocation, which mitigates execution-related risks.
- [COMMAND_EXECUTION]: The skill references the
markdownlint-cli2command for structural auditing. This is a standard linting utility and the usage pattern provided is for manual or guided reference only. - [INDIRECT_PROMPT_INJECTION]: The skill defines patterns for auditing external markdown content, establishing a theoretical ingestion surface for untrusted data.
- Ingestion points: External markdown documents audited for accessibility compliance.
- Boundary markers: No specific boundary markers or "ignore" instructions are defined within this reference module.
- Capability inventory: The skill is strictly reference data; it has no network access, file-writing scripts, or tool-invocation capabilities.
- Sanitization: Fix templates provide structural remediation for accessibility issues (e.g., adding alt text) but do not perform general input sanitization.
Audit Metadata