circleci-deploy-setup
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
circleciCLI andgitto perform auditing of the local repository, list existing pipelines, and register new pipeline definitions. These operations are essential for the primary functionality of the skill. - [INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection attack surface because it reads and processes repository files (e.g., config files, deployment scripts, Helm charts, and Dockerfiles) to infer deployment parameters.
- Ingestion points: The agent reads
.circleci/config.yml,scripts/deploy.sh, Helm values, Dockerfiles, and Makefiles into its context. - Boundary markers: The skill instructs the agent to show all inferred values to the user for correction before proceeding.
- Capability inventory: The skill can perform file system writes and call CircleCI API creation endpoints through the CLI.
- Sanitization: The skill includes a detailed verification suite with six invariants to check generated configurations and mandates a user-approved diff review before committing any changes.
- [EXTERNAL_DOWNLOADS]: The skill references official service endpoints for CircleCI and well-known monitoring providers like Datadog and Grafana for release validation and webhook setup. These are standard integrations required for the skill's deployment monitoring features.
Audit Metadata