workflow-ship-and-observe

Pass

Audited by Gen Agent Trust Hub on Oct 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and process data from external, potentially untrusted sources.
  • Ingestion points: The agent is directed to read configuration files such as vercel.json, netlify.toml, and Dockerfile (Phase 0), as well as monitoring data from Sentry, provider logs, and Supabase logs (Phase 4).
  • Boundary markers: The instructions lack explicit delimiters or warnings to ignore instructions embedded within the logs or metadata being read.
  • Capability inventory: The agent is tasked with executing deployment CLI commands, running smoke checks via tools like Playwright, and writing state to local files (.cursor/ship-state.md).
  • Sanitization: No sanitization or validation protocols are defined for handling content retrieved from external monitoring endpoints or logs.
  • [COMMAND_EXECUTION]: The skill involves the execution of various deployment and verification commands.
  • Evidence: The instructions involve running commands such as vercel deploy --prod, flyctl (implied), and infrastructure-specific CLI tools to trigger pipelines and verify deployments.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 7, 2026, 11:27 PM