methodology-work

Pass

Audited by Gen Agent Trust Hub on Aug 31, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process and ingest external data, such as user corrections and feedback from other agents, to update the methodology knowledge graph. This creates a surface where malicious instructions embedded in the feedback could influence the agent's behavior when modifying the graph.
  • Ingestion points: Processes external user feedback and agent logs as described in the sections for recording principles and processing feedback in SKILL.md.
  • Boundary markers: Absent; the skill does not define specific delimiters or instructions to isolate the ingested feedback from the agent's operational instructions.
  • Capability inventory: Employs specialized tools for interacting with the methodology realm (iskron_orient, iskron_search, iskron_look, and iskron_semantic_search) as detailed in SKILL.md.
  • Sanitization: Absent; no specific sanitization or validation logic is defined for the content of the feedback before it is recorded into the system.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 31, 2026, 07:38 AM
Security Audit — agent-trust-hub — methodology-work