skills/bbenefield89/skills/deliver/Gen Agent Trust Hub

deliver

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data including tickets, specifications, and repository code which are interpolated into the agent's context. A malicious actor could embed instructions within these sources to influence the agent's actions during implementation or review.\n
  • Ingestion points: Tickets, specifications, and repository files (references/preflight.md, references/review.md).\n
  • Boundary markers: The skill relies on separate workflow phases but does not explicitly use structural delimiters for untrusted content.\n
  • Capability inventory: The agent has the ability to modify files and execute shell commands for testing and validation.\n
  • Sanitization: The instructions do not specify any filtering or sanitization of external data.\n- [COMMAND_EXECUTION]: The skill is instructed to discover and run validation commands from repository manifests, build configurations, and CI scripts. This is a functional requirement for delivery but involves executing logic defined in the local environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 12:37 AM
Security Audit — agent-trust-hub — deliver