The Agent Skills Directory

[SAFE]: The skill provides a legitimate, structured workflow for software maintenance and issue tracking. All file operations are constrained to the local .codestable directory.
[COMMAND_EXECUTION]: The skill references internal project scripts cs-onboard and search-yaml.py. These appear to be part of the skill's infrastructure for environment configuration and documentation indexing. No evidence of unsafe command construction or injection was found.
[PROMPT_INJECTION]: The skill processes user descriptions of software issues, creating an indirect prompt injection surface. Ingestion points: User-supplied bug reports and local markdown files in .codestable/issues/. Boundary markers: None present in the routing logic. Capability inventory: File system globbing and routing to sub-skills. Sanitization: No explicit sanitization or input validation is performed at this layer.

cs-issue