nw-research-methodology

Pass

Audited by Gen Agent Trust Hub on Apr 4, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is designed to process external research data, creating a surface for indirect prompt injection where malicious instructions could be embedded in the sources being analyzed. \n
  • Ingestion points: External research sources, URLs, and academic documents identified during the research phase. \n
  • Boundary markers: The methodology utilizes structured markdown templates to organize findings, providing logical separation between data and instructions. \n
  • Capability inventory: The skill instructs the agent to write files to the local file system, including the skill configuration directory. \n
  • Sanitization: Includes quality gates such as source reputation scores and mandatory cross-referencing to validate external content. \n- [COMMAND_EXECUTION]: The skill instructions direct the agent to write new skill files to the ~/.claude/skills/ directory. Writing to the agent's internal configuration paths is a high-privilege action that could be used to alter agent behavior if the source information is compromised.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 4, 2026, 03:43 PM
Security Audit — agent-trust-hub — nw-research-methodology