The Agent Skills Directory

[Indirect Prompt Injection] (HIGH): The skill is highly vulnerable to indirect prompt injection due to its operational capabilities. * Ingestion points: Untrusted data enters via CIRCLECI_GET_TEST_METADATA (test messages) and CIRCLECI_GET_JOB_ARTIFACTS (file content). * Boundary markers: None are specified in the instructions to separate external data from system instructions. * Capability inventory: The skill can execute CIRCLECI_TRIGGER_PIPELINE, allowing an injection to trigger unauthorized builds. * Sanitization: No sanitization or validation of external data is mentioned.
[External Downloads] (MEDIUM): The skill requires the configuration of an external MCP server at https://rube.app/mcp. This domain is not a trusted source, meaning the agent's tool logic is provided by an unverified third party.
[Data Exfiltration] (LOW): The skill facilitates the retrieval of sensitive CI/CD artifacts and pipeline configurations, which could lead to organizational data exposure if the agent's context is compromised.

circleci-automation