keyapi-facebook

Warn

Audited by Gen Agent Trust Hub on Jul 14, 2026

Risk Level: MEDIUMDATA_EXFILTRATIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION]: The script scripts/configure-keyapi-auth.mjs accesses and reads sensitive system files in the user's home directory, specifically shell profiles such as ~/.bashrc, ~/.zshrc, .profile, and Windows PowerShell profiles, to detect or retrieve the KEYAPI_TOKEN credential.
  • [COMMAND_EXECUTION]: The script scripts/configure-keyapi-auth.mjs establishes persistence for credentials by modifying the user's shell environment files to export the KEYAPI_TOKEN variable across sessions.
  • [EXTERNAL_DOWNLOADS]: The skill performs automated network requests to docs.keyapi.ai to fetch documentation and api.keyapi.ai to execute Facebook data operations.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface. Evidence: (1) Ingestion point: scripts/search-keyapi-docs.mjs fetches external content from https://docs.keyapi.ai/llms.txt. (2) Boundary markers: No delimiters or ignore instructions are present. (3) Capability inventory: The skill can execute shell commands via local scripts and perform network requests. (4) Sanitization: The ingested documentation data is used to dynamically construct API requests without validation or sanitization.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jul 14, 2026, 09:25 AM
Security Audit — agent-trust-hub — keyapi-facebook