workflow-e2e
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute a local maintenance script
node .claude/scripts/lib/workflow-skill-contract.cjs --writeto update workflow configurations. - [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection due to the ingestion of untrusted data combined with significant operational capabilities.
- Ingestion points: The workflow processes external
recording JSONfiles, product source code, and API specifications (e.g., in the Source Dispatch section). - Boundary markers: Absent. The skill does not define clear delimiters or use 'ignore' instructions for the data it processes.
- Capability inventory: The skill can execute shell commands (test runners), run local scripts, write reports to the filesystem (under
tmp/reports/), and spawn sub-agents. - Sanitization: Absent. No evidence of content sanitization or validation is present in the instructions.
Audit Metadata