completion-verification

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to run project-level commands and evaluate their output to ensure work is complete. This creates a potential surface for indirect prompt injection where malicious instructions embedded in a project's build or test output could influence the agent's actions.
  • Ingestion points: Reading the output of commands executed during verification steps (SKILL.md).
  • Boundary markers: Absent; the skill does not instruct the agent to use delimiters or ignore embedded instructions within command output.
  • Capability inventory: The skill assumes the agent has capabilities to execute shell commands and read the project file system.
  • Sanitization: Absent; no escaping or validation of the external command output is suggested before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 01:17 PM
Security Audit — agent-trust-hub — completion-verification