add-llm-provider

Pass

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read external data from system logs to determine configuration parameters, creating a potential surface for indirect injection if the logs contain untrusted content.\n
  • Ingestion points: `docker exec awf-squid cat /var/log/squid/access.log` referenced in Step 1.\n
  • Boundary markers: None specified for the log content.\n
  • Capability inventory: Creating and modifying JavaScript/TypeScript source files, updating JSON network policies, and executing Docker/CLI commands.\n
  • Sanitization: None specified; the agent is expected to manually inspect and interpret the log content.\n- [COMMAND_EXECUTION]: The skill provides templates for executing Docker commands and the awf CLI to inspect and configure the environment.\n
  • Evidence: `docker exec awf-squid cat /var/log/squid/access.log` and `awf --allow-domains api2.cursor.sh,api3.cursor.sh -- cursor-agent ...` snippets in SKILL.md.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 11, 2026, 08:33 AM
Security Audit — agent-trust-hub — add-llm-provider