The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute git commands (status, diff, add, commit) and runs a pre-commit hook script that invokes local project utilities such as npm, bun, pnpm, yarn, make, ruff, and eslint.
[DYNAMIC_EXECUTION]: The pre-commit-lint.sh script dynamically identifies and executes linting commands defined in local project files like package.json and Makefile, enabling the execution of arbitrary scripts defined within the workspace.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from git diff and status outputs, creating a vulnerability surface where malicious code comments or diff content could influence the agent's behavior.
Ingestion points: Git status and git diff outputs are read into the agent's context during the workflow defined in SKILL.md.
Boundary markers: The instructions lack explicit delimiters or boundary markers to differentiate between diff content and agent instructions.
Capability inventory: The skill has access to bash (git), read, write, and edit tools, which could be exploited if an injection is successful.
Sanitization: No sanitization, escaping, or validation of the diff output is performed before it is processed by the model.

git-commit