The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and analyzes untrusted data from external sources.
Ingestion points: Commit messages retrieved via git log, code changes via git diff, and various project configuration files (identified in SKILL.md).
Boundary markers: There are no explicit boundary markers or instructions to ignore embedded commands when the agents analyze external text.
Capability inventory: The skill utilizes Write and Edit tools for file modification and Bash for git operations (git add, git commit) (SKILL.md).
Sanitization: No sanitization or content validation is performed on the commit messages or code content before they are processed by the analysis agents.
[COMMAND_EXECUTION]: The skill uses the Bash tool to execute system-level git commands.
Evidence: The skill is allowed to run git diff, git status, git log, git add, and git commit. While these are restricted to specific git subcommands, they represent significant system interaction required for the skill's primary purpose.

bump-version