The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes repository-wide data, including git history and directory structures, which can be influenced by external actors through commit messages or filenames.\n
Ingestion points: Data is ingested via git diff and git describe in scripts/detect-changes.sh, and through grep searches across the repository root in SKILL.md.\n
Boundary markers: The skill does not provide delimiters or instructions to the agent to treat data from the repository as untrusted or to ignore embedded instructions.\n
Capability inventory: The skill utilizes the Bash, Write, and Edit tools to modify repository files and execute scripts (SKILL.md).\n
Sanitization: No sanitization or validation is applied to the output of git commands or file names before they are used to drive the versioning logic.\n- [COMMAND_EXECUTION]: The skill executes shell scripts and git commands to perform version management and repository analysis.\n
Evidence: Execution of detect-changes.sh and bump-version.sh via the Bash tool (SKILL.md).\n
Evidence: Usage of git, find, grep, sed, and cut within scripts/detect-changes.sh to analyze changes.\n- [DYNAMIC_EXECUTION]: The skill executes internal scripts using paths determined at runtime from environment variables.\n
Evidence: Shell scripts are invoked from ${CLAUDE_PLUGIN_ROOT}/skills/edit-plugin/scripts/ (SKILL.md).

edit-plugin