The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied code (files, functions, or modules) which could contain adversarial instructions designed to influence the agent's behavior during the audit process.
Ingestion points: The 'Input & Boundary' section of SKILL.md specifies that the agent receives files, functions, modules, or repository scopes for inspection.
Boundary markers: There are no instructions for using delimiters or specific 'ignore' directives when processing the audited content.
Capability inventory: The skill is capable of staging changes for external tools via /forge and routing decisions to /clarify or /architect-decision.
Sanitization: No sanitization, filtering, or escaping of the ingested code content is specified in the execution protocol.

simplify-audit