The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructions specify a fallback mechanism to fetch shared reference files from the author's GitHub repository (github.com/levnikolaevich/claude-code-skills) if the required files are missing from the local environment.
[COMMAND_EXECUTION]: The workflow includes pre-flight checks that verify the environment by executing commands to check for the availability and versions of 'docker' and 'docker compose'.
[REMOTE_CODE_EXECUTION]: The skill is designed to coordinate the execution of build scripts, test suites, and container deployments by delegating to specialized worker skills. While the coordinator does not execute these directly, the overall system is intended to run project-specific code and containers.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it processes external project data and test results.
Ingestion points: Reads project configuration files (package.json, *.csproj, docker-compose.yml) and aggregates logs/results from build and test phases.
Boundary markers: No explicit delimiters or instruction-ignore warnings are used for interpolated data.
Capability inventory: Delegates tasks to skills capable of arbitrary code execution (builds, tests, and container management).
Sanitization: No evidence of sanitization or validation for the ingested project data or aggregated results.

ln-780-bootstrap-verifier