The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the git CLI for local repository operations, including branching, staging, committing changes, and pushing to remotes. It also uses the gh CLI to interact with GitHub for creating pull requests, viewing PR URLs, and adding review comments.
[REMOTE_CODE_EXECUTION]: Implements a dynamic workflow using the platform's native JavaScript orchestration capabilities to manage a pipeline of sub-agents that perform specialized tasks like logic review, security analysis, and code patching.
[DATA_EXFILTRATION]: Facilitates the transfer of repository contents and generated PR metadata to the user's configured Git remote (e.g., GitHub). These operations are transparent, documented, and required for the skill's primary function of shipping code.
[EXTERNAL_DOWNLOADS]: The documentation provides instructions for installing the skill from the author's public repository (tylergibbs1/dispatch) using the standard skills CLI installation method.
[PROMPT_INJECTION]: The skill processes untrusted data from PR diffs, creating an indirect prompt injection surface.
Ingestion points: Code diffs are retrieved via git diff in references/review-fix-workflow.md and passed directly to review agents.
Boundary markers: The instructions lack explicit delimiters to separate the diff content from agent instructions, relying on the agents to interpret the context correctly.
Capability inventory: The workflow has the ability to modify the local file system (applying fixes) and perform network pushes to the git remote.
Sanitization: The skill mitigates risks through a multi-agent adversarial verification phase where findings must be confirmed by independent skeptics before any automated fixes are performed.

shipping-pr-reviews