The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from local project files to generate a metrics plan, representing a low-risk indirect injection surface.
Ingestion points: Project PRDs and CLAUDE.md files are read in Step 1 and Step 2 to determine project context and stack.
Boundary markers: There are no explicit instructions or delimiters used when interpolating content from project files into the generated metrics plan.
Capability inventory: The skill is restricted to Read, Grep, Glob, Write, and AskUserQuestion tools, and it specifically writes its output to docs/metrics-plan.md via the Write tool.
Sanitization: No explicit sanitization or filtering of external project content is mentioned before it is processed by the agent.
[SAFE]: No hardcoded credentials, API keys, or secrets were detected in the skill instructions or reference files.
[SAFE]: No obfuscation techniques, such as Base64 encoding of commands or hidden Unicode characters, were found in the provided files.
[SAFE]: The skill does not perform any network operations or remote code execution; it focuses on local project analysis and documentation generation.

solo-metrics-track