paperclip-create-agent

Pass

Audited by Gen Agent Trust Hub on Apr 8, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses curl to interact with the Paperclip API for administrative tasks, such as fetching configuration templates and submitting new agent hire requests. This behavior is consistent with the skill's documented purpose.
  • [SAFE]: Credential management follows best practices by using environment variables ($PAPERCLIP_API_KEY) for authentication. All network traffic is directed to the platform's designated API endpoint.
  • [PROMPT_INJECTION]: The skill ingests external data from the platform's configuration guides and user-generated approval comments, which theoretically allows for indirect prompt injection. However, this is a standard operational requirement for the tool's functionality.
  • Ingestion points: Fetches data from llms/agent-configuration.txt, llms/agent-configuration/claude_local.txt, and /api/approvals/<approval-id>/comments.
  • Boundary markers: None.
  • Capability inventory: Execution of subprocesses via curl commands.
  • Sanitization: None identified for API-fetched text inputs.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 8, 2026, 02:29 PM
Security Audit — agent-trust-hub — paperclip-create-agent