autogpt-agents
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The platform is designed to ingest and process data from external sources, creating a surface for potential indirect prompt injection.
- Ingestion points: API endpoints for manual execution and webhooks defined in SKILL.md.
- Boundary markers: The provided examples in SKILL.md do not explicitly show boundary markers for all block inputs.
- Capability inventory: The platform supports LLM generation, decision logic, and integrations with external services (GitHub, Google, Discord, Notion) as described in SKILL.md and references/advanced-usage.md.
- Sanitization: Documentation in references/advanced-usage.md mentions the use of sanitized inputs in error tracking contexts.
- [COMMAND_EXECUTION]: The documentation provides standard CLI commands for system administration and project setup.
- Evidence: SKILL.md and references/troubleshooting.md include commands for Docker management (docker compose up), package management (npm install, poetry run), and system service management (sudo systemctl start docker).
- [EXTERNAL_DOWNLOADS]: The skill guides users to download the platform source code from its official repository.
- Evidence: SKILL.md contains instructions to clone the project from https://github.com/Significant-Gravitas/AutoGPT.git.
Audit Metadata