The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local commands such as git, node, yarn, rg, and grep. These are used to determine project state, run maintenance scripts (e.g., scripts/check_changes.ts, scripts/lint_ts_projects.js), and run tests. This behavior is confined to the local repository environment and is consistent with developer workflow automation.\n- [PROMPT_INJECTION]: The skill ingests data from local project files and command outputs (Indirect Prompt Injection surface). Ingestion points: kibana.jsonc, tsconfig.json, and git diff output. Boundary markers: None. Capability inventory: Local shell execution of project-specific scripts. Sanitization: None. Given the specific project context and the use of the disable-model-invocation flag, this is considered a safe application of the capability.

branch-readiness-checks