The Agent Skills Directory

[COMMAND_EXECUTION]: The skill performs automated checks by invoking shell commands including git, gh (GitHub CLI), cargo, and cargo-nextest. These commands are used to inspect repository history, query GitHub Actions status, and run the project's test and linting suites. The commands are specific to the Rust ecosystem and the GitHub platform.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests and processes data from external, potentially untrusted sources within the repository.
Ingestion points: The agent reads the contents of TODO.md, files within the todo/ directory, and SECURITY_EXCEPTIONS.md. It also captures output from git log (commit messages) and gh pr list (PR metadata).
Boundary markers: The instructions do not define specific delimiters or explicit 'ignore' instructions to prevent the agent from accidentally following instructions that could be embedded within the audited files or commit messages.
Capability inventory: The agent has the capability to execute shell commands via git, gh, and cargo tools within the local repository environment.
Sanitization: There is no evidence of content sanitization or validation for the data retrieved from repository files or command outputs before it is interpolated into the agent's context.

jackin-release-check