commissaire-aux-comptes

Pass

Audited by Gen Agent Trust Hub on Apr 19, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection as it processes data from external files that could contain malicious instructions designed to subvert the agent's behavior.\n
  • Ingestion points: The skill instructions in 'SKILL.md' direct the agent to read and analyze 'company.json' and 'data/fec-*.txt'.\n
  • Boundary markers: No instructions are provided to the agent to treat data from these files as untrusted or to isolate the file content from the skill's operational instructions.\n
  • Capability inventory: The skill uses the agent's analytical and reporting capabilities to process user-supplied files; these capabilities could be misused if instructions are embedded in the ingested data.\n
  • Sanitization: There are no procedures defined in the skill for validating or sanitizing the content of accounting files before analysis.\n- [NO_CODE]: The skill does not contain any executable scripts (Python, Node.js, or shell); all logic is provided as natural language instructions, which significantly reduces the technical attack surface.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 19, 2026, 12:48 PM