The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local shell script (scripts/ship-check.sh) to run project-specific development tools such as npm, yarn, pnpm, bun, and ruff. These executions are transparently documented and align with the skill's purpose as a pre-ship verification tool.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface (Category 8) by ingesting and analyzing untrusted content from the local repository.
Ingestion points: The agent processes data from git diff, git status, git log, and project files like package.json and pyproject.toml to generate summaries and commit messages.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to distinguish between its instructions and the potentially adversarial natural language content within the code diffs being reviewed.
Capability inventory: The agent has the ability to write reports to the filesystem and execute git commit and git push commands upon user confirmation.
Sanitization: The skill does not implement sanitization or escaping of the ingested diff content before it is processed by the agent.

ship-check