The Agent Skills Directory

[Command Execution] (SAFE): The skill utilizes the Bash tool to execute git log -5. This is a non-privileged, read-only command used to provide context on recent project activity. No arbitrary command injection vectors were found.
[Indirect Prompt Injection] (LOW): The skill reads several project-related files which could potentially contain malicious instructions if the repository environment is untrusted.
Ingestion points: ideas/{project}/README.md, ideas/{project}/project-brief.md, ideas/{project}/critique.md, CLAUDE.md, and about-me.md via the Read tool.
Boundary markers: Absent. The skill does not use specific delimiters or instructions to ignore potential commands within the files being read.
Capability inventory: The skill has access to file reading (Read, Glob, Grep) and shell command execution (Bash).
Sanitization: Absent. Content from external files is passed directly to the model's context without sanitization.

sanity-check