wisdom-accountability-coach

Pass

Audited by Gen Agent Trust Hub on Mar 7, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its core functionality of reading external data and writing to the local file system.
  • Ingestion points: The agent ingests data from untrusted sources via the Read tool (local files), WebFetch (URLs), and search tools (mcp__firecrawl__firecrawl_search, mcp__brave-search__brave_web_search).
  • Boundary markers: The instructions do not define boundary markers or 'ignore' instructions to isolate data from the system prompt, increasing the risk that embedded instructions in processed data could be executed.
  • Capability inventory: The skill allows the use of Write, Edit, and TodoWrite tools, which enables the modification of files based on potentially malicious instructions found in ingested data.
  • Sanitization: There is no evidence of input validation or sanitization logic to verify the integrity of external content before it is processed or written to storage.
  • [NO_CODE]: No executable code, scripts, or binaries are included in the skill package; it consists entirely of markdown-based instructions and reference materials.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 7, 2026, 01:29 PM