The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted document content which could contain malicious instructions designed to influence the agent's behavior during the review process.
Ingestion points: The skill reads <STEM>.revN.md files from the docs/ directory as its primary input.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat the document content as data rather than instructions.
Capability inventory: The skill utilizes git commands and shell execution for polling, providing a surface for potentially exploited capabilities.
Sanitization: No sanitization or validation of the input document content is specified.
[COMMAND_EXECUTION]: The skill relies on shell commands and local script execution to manage the review loop.
Evidence: Executes Git operations including git fetch, git log, git add, and git commit to manage state.
Evidence: Generates and runs a bash script to poll the file system for changes in document file size.
[EXTERNAL_DOWNLOADS]: The skill interacts with external repositories to verify branch state.
Evidence: Uses git fetch to synchronize the local environment with a remote integration branch.

review-doc-loop