The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill's BUILD phase explicitly instructs the agent to write functional checks and implementation code to the local filesystem and then execute them to verify the build.
[COMMAND_EXECUTION]: The agent is directed to run the /serve command, which initiates a local web server (defaulting to port 4399) to facilitate decision reviews through decisions.html.
[INDIRECT_PROMPT_INJECTION]: The skill demonstrates a vulnerability surface for indirect prompt injection as its operational logic is governed by the content of external markdown and HTML files.
Ingestion points: The agent reads instructions and state from docs/roadmap.md, docs/_current.md, _milestone.md, and decisions.md to determine its next actions.
Boundary markers: No specific boundary markers or instructions to ignore embedded commands within the ingested data were identified.
Capability inventory: The skill possesses the ability to write files, execute shell commands via /serve, and run arbitrary code during the BUILD process.
Sanitization: There is no evidence of sanitization or validation of the content read from the project documentation before it influences agent behavior.

next