tailor-core
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill framework is designed to ingest and process external data from a project's repository, which could potentially contain malicious instructions intended to influence the agent's behavior.
- Ingestion points: The operational contract (Section 3: Progressive Disclosure) explicitly instructs the agent to read various markdown files from the
.ai/directory (e.g.,CONSTITUTION.md,DATABASE.md,API.md) and thespecs/directory to gain context. - Boundary markers: The instructions do not define specific delimiters or "ignore embedded instructions" warnings for the content of these external files.
- Capability inventory: The skill possesses significant capabilities, including routing commands to modify production source code via
/spec-implement, performing security audits, and managing project dependencies. - Sanitization: There are no documented procedures for sanitizing, validating, or escaping the content read from the
.ai/orspecs/files before it is processed by the agent.
Audit Metadata