The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it fetches and processes externally controlled data from GitHub without explicit boundary markers or sanitization.
Ingestion points: Data is ingested from GitHub PR bodies and titles via gh pr view and release descriptions via gh release view in Step 1 and Step 2.
Boundary markers: There are no instructions to the agent to treat the fetched content as data or to ignore any embedded instructions within that content.
Capability inventory: The skill utilizes shell commands (gh, git, grep, python3), writes to local HTML files (site/changelog.html), and launches a local web server for verification.
Sanitization: The skill lacks validation or filtering mechanisms for the content fetched from GitHub before it is processed by the agent.
[COMMAND_EXECUTION]: The skill performs various shell operations to interact with the repository and verify changes.
Uses gh CLI to fetch release and pull request data.
Uses grep and git diff to locate markers and verify changes in site/changelog.html.
Launches a local server using python3 -m http.server for visual verification of the generated changelog.

changelog