The Agent Skills Directory

[COMMAND_EXECUTION]: The skill directs the agent to execute shell-level operations using git and the GitHub CLI (gh) for repository management, including branching, tagging, and pull request interactions. It also instructs the agent to run project-specific tools and validation gates defined within the target repository.
[INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted data from the repository's history and metadata to determine release versions and content.
Ingestion points: The agent reads conventional commit messages, project changelogs, and workflow configuration files from the local repository.
Boundary markers: The instructions do not define explicit delimiters or security warnings to be used when interpolating repository data into the agent's context.
Capability inventory: The agent can modify files, execute Git commands, create GitHub pull requests, and trigger remote workflows via the GitHub CLI.
Sanitization: There are no specified procedures for sanitizing or validating inputs derived from commit messages or other repository-based data sources.
[EXTERNAL_DOWNLOADS]: The skill requires active network access to synchronize with remote Git repositories and interact with the GitHub API for triggering workflows and monitoring release status.

create-release