The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes several standard command-line utilities, including gh, glab, grep, jq, and wc, to retrieve and inspect pull request metadata, file diffs, and project structures.
[DYNAMIC_EXECUTION]: The workflow involves running local test suites via npm test and pytest. This is used to evaluate code coverage changes and ensure functional correctness of the pull request being analyzed.
[DATA_EXFILTRATION]: The skill performs network requests to well-known service endpoints for Jira (atlassian.net) and Linear (api.linear.app) to verify issue status. It implements secure credential handling by passing tokens through curl via stdin (-K -) or recommending the use of ~/.netrc, which prevents sensitive information from appearing in process lists or shell history.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external, untrusted content from pull request diffs and descriptions.
Ingestion points: Data enters the agent's context through gh pr diff, gh pr view, glab mr diff, and glab mr view commands (SKILL.md).
Boundary markers: The instructions do not define specific delimiters to separate the untrusted diff content from the analysis instructions.
Capability inventory: The skill has access to shell execution (grep, jq), local test execution (npm test, pytest), and network communication (curl).
Sanitization: The skill employs targeted grep patterns for analysis but lacks explicit sanitization of the untrusted diff input.

pr-review-expert