The Agent Skills Directory

[SAFE]: The skill's primary behavior involves reading and analyzing local project source files and a specific configuration file (.tt-impeccable.md). These operations are performed within the local project context and are necessary for the skill's stated purpose of design critique. No unauthorized file access or modification was observed.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it processes untrusted data from project source code. Ingestion points: project source files and the .tt-impeccable.md configuration file; Boundary markers: absent; Capability inventory: limited to local file reading and generating a text-based report; Sanitization: absent. Despite the lack of sanitization, the risk is negligible because the skill does not execute code, perform network requests, or modify the system based on the ingested content.

tt-critique