The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes standard development tools including git, gh (GitHub CLI), and pnpm. These are used to analyze branch history, retrieve pull request metadata, and verify repository state. All commands are executed via execFileSync or spawnSync, which avoids shell injection vulnerabilities by passing arguments as a discrete array.
[EXTERNAL_DOWNLOADS]: The script scripts/preflight-changelog-ci.mjs includes a command to run pnpm install --frozen-lockfile. This is an optional utility meant for CI environments to ensure dependencies are installed from the official registry. This targets a well-known service and is documented as part of the skill's developer tooling.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources such as git commit messages, branch names, and GitHub PR data. While this represents an attack surface for indirect prompt injection, the risk is minimal as the skill primarily uses this data for structured metadata extraction (e.g., issue IDs, categories) and uses regex-based masking to prevent data corruption during processing. In accordance with guidelines, this surface is documented with a low severity assessment.

changelog