The Agent Skills Directory

[COMMAND_EXECUTION]: The skill employs standard Rust toolchain commands and shell utilities (cargo, git, gh, jq, grep, sed) to manage project state, verify code quality, and automate release tasks. Key operations include cargo check, cargo update, and cargo build.
[DATA_EXFILTRATION]: The skill performs expected network operations to push release branches to remote repositories and create pull requests on GitHub via git push and gh pr create.
[PROMPT_INJECTION]: The instructions include a directive for the agent to avoid referencing AI tools or co-authorship in commit messages and pull request descriptions. This is a common stylistic requirement for professional software projects.
[PROMPT_INJECTION]: The skill ingests data from local project files, such as Cargo.toml and CHANGELOG.md, to determine version information and release notes. This constitutes a potential surface for indirect prompt injection if those files contain malicious instructions, though the impact is scoped to the release automation workflow.
Ingestion points: Cargo.toml (parsed for version metadata) and CHANGELOG.md (parsed for unreleased changes).
Boundary markers: No explicit delimiters are specified for separating file content from the agent's instructions during pull request body generation.
Capability inventory: File system access, shell execution, and remote repository management via Git and GitHub CLI.
Sanitization: No specific validation or escaping is applied to the content extracted from project files before it is processed.

rust-release