circleci-automation

Pass

Audited by Gen Agent Trust Hub on Jul 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill instructions and workflows align with the stated purpose of automating CircleCI operations. No evidence of malicious intent, data exfiltration of sensitive local files, or obfuscation was found.
  • [EXTERNAL_DOWNLOADS]: The skill directs users to add https://rube.app/mcp as an MCP server. This is a legitimate integration point for the Composio 'Rube' platform, a well-known service for AI toolkits.
  • [PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it processes data from external CI/CD logs and metadata.
  • Ingestion points: Untrusted data enters the agent context through CIRCLECI_GET_TEST_METADATA and CIRCLECI_GET_JOB_ARTIFACTS (SKILL.md).
  • Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within CI outputs.
  • Capability inventory: The skill allows triggering pipelines (CIRCLECI_TRIGGER_PIPELINE) and viewing configurations (CIRCLECI_GET_PIPELINE_CONFIG) (SKILL.md).
  • Sanitization: No evidence of output sanitization or filtering of CircleCI logs is present.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 18, 2026, 04:24 PM
Security Audit — agent-trust-hub — circleci-automation