generate-eou-candidates

Pass

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious patterns, obfuscation, or unauthorized network activity detected. The skill's behavior aligns with its documented purpose of structured workflow analysis.
  • [COMMAND_EXECUTION]: The skill uses the Read and Write tools to access project registries and save analysis results within the foundry/ directory. These operations are scoped to the local project environment.
  • [PROMPT_INJECTION]: The skill processes untrusted content from the $workflow input, which represents a surface for indirect prompt injection.
  • Ingestion points: Data enters via the $workflow parameter in SKILL.md.
  • Boundary markers: No explicit markers are used to isolate user-provided workflow text from the skill's logic.
  • Capability inventory: The skill employs Read, Write, and Grep tools to summarize inputs and produce artifacts (SKILL.md).
  • Sanitization: The skill does not perform specific sanitization of the input text, though it follows a strict output schema.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 23, 2026, 06:08 PM
Security Audit — agent-trust-hub — generate-eou-candidates