The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute local development tools such as pytest, ruff, and mypy using the uv package manager.
[COMMAND_EXECUTION]: Includes instructions to run local Python scripts within the repository, specifically scripts/count_mcp_tools.py and scripts/check_cline_skills.py.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by instructing the agent to ingest and follow instructions from external files in the .clinerules/ and .claude/skills/ directories.
Ingestion points: Files located at .clinerules/workflows/, .claude/skills/pubmed-*, and .claude/skills/pipeline-persistence in the local repository.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the skill definition for these external sources.
Capability inventory: The skill allows subprocess execution and local command running via uv run and python3 across various development scripts.
Sanitization: There is no evidence of sanitization or validation of the content within the referenced external workflow files.

pubmed-search-mcp-harness