paperclip-create-agent
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external API endpoints, including documentation files and user-provided board feedback, which are used to configure new agents.
- Ingestion points: API responses from
/llms/agent-configuration.txt,/llms/agent-icons.txt, and/api/approvals/<approval-id>/comments(SKILL.md). - Boundary markers: Absent. The instructions do not specify the use of delimiters or instructions to ignore embedded commands within the ingested content.
- Capability inventory: The skill uses
curlto perform network operations, including POST requests that create new agents with defined capabilities and prompts (SKILL.md). - Sanitization: No sanitization or validation logic is defined for the data retrieved from the API before it influences the agent's workflow.
- [COMMAND_EXECUTION]: The skill utilizes
curlshell commands to interact with the Paperclip service. These commands handle sensitive authentication tokens via environment variables to perform administrative actions like agent creation and comment submission. - [EXTERNAL_DOWNLOADS]: The skill references a configuration example for a skill hosted in Vercel Labs' public repository.
Audit Metadata