The Agent Skills Directory

[COMMAND_EXECUTION]: The skill makes extensive use of shell commands and internal agent functions to perform development tasks.
Executes Git operations including branching, committing, and cherry-picking.
Runs build and test suites using npm and docker.
Orchestrates releases using the gh (GitHub CLI) for creating releases, uploading artifacts, and managing Pull Requests.
[EXTERNAL_DOWNLOADS]: The skill triggers downloads and execution of external tools and packages.
Uses npx to download and execute claude-flow (specifically the @alpha version) from the npm registry.
Interacts with the GitHub API and npm registry for package publication and data retrieval.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data that could be controlled by an attacker to influence the agent's behavior.
Ingestion points: Fetches commit messages and Pull Request titles/labels using gh api and gh pr list in SKILL.md.
Boundary markers: The skill lacks explicit boundary markers or instructions to the LLM to ignore potentially malicious instructions embedded within the ingested commit data.
Capability inventory: The agent has high-privilege capabilities including file system writes, shell command execution, and network publication (npm/docker/GitHub).
Sanitization: There is no evidence of sanitization or filtering of the fetched commit/PR text before it is used to generate release notes or changelogs.

github-release-management