The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local Python script via the command uv run research/knowledge-explorer.py. This is used to filter research entries based on layer metadata and is a functional component of the evaluation process.
[EXTERNAL_DOWNLOADS]: The skill's documentation references an external GitHub repository (github.com/Jamie-BitFlight/sam-flow-experiments) for running flow experiments. This repository is owned by the skill's author and represents a vendor resource.
[INDIRECT_PROMPT_INJECTION]: The skill systematically reads and validates project documentation and research files, creating an attack surface where malicious content in those files could influence the agent's report or the fixes it applies.
Ingestion points: Documentation files located in plugins/development-harness/docs/ and various research markdown files.
Boundary markers: No specific delimiters or safety instructions are used to isolate the ingested file content from the agent's logic.
Capability inventory: The skill can execute local scripts via uv and has the capability to modify files when the --fix argument is provided.
Sanitization: There is no evidence of sanitization or filtering of the content read from external files before it is processed by the agent.

evaluate-sdlc-layers