using-skilder
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill outlines a process for fetching and acting upon external data such as roles, skills, and files which may contain embedded instructions.
- Ingestion points: Data enters the agent's context through the learn and open_app tools from the remote workspace endpoint (app.skilder.ai).
- Boundary markers: Instructions explicitly tell the agent to treat App content as information and not instructions to mitigate the risk of command injection.
- Capability inventory: The agent can invoke high-privilege tools including run_write_tool and execute_script based on the ingested schemas.
- Sanitization: The skill relies on server-side guardrails and session-based access control rather than client-side sanitization.
Audit Metadata