Skills
Navigate
skills/swan-gtm/gtm-skills/clear-my-desk/Gen Agent Trust Hub

clear-my-desk

Pass

Audited by Gen Agent Trust Hub on Jul 28, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external task data and sequence information which may contain adversarial content designed to influence the agent's triage logic.
  • Ingestion points: Data is ingested via swan-search-user-tasks and swan-search-sequences (SKILL.md).
  • Boundary markers: The skill does not define explicit boundary markers or 'ignore' instructions for the processed data.
  • Capability inventory: The skill can execute Python code via swan-execute-code and perform task mutations via swan-acknowledge-user-task.
  • Sanitization: No explicit sanitization or filtering of task content is described before classification.
  • [COMMAND_EXECUTION]: The skill uses swan-execute-code to dynamically generate and run Python scripts (using pandas) for triaging large numbers of items and batching approval calls.
  • Evidence: Instructions in Step 2 and Step 5 describe generating code to process JSON data and execute batches of API calls.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 28, 2026, 04:39 PM
Security Audit — agent-trust-hub — clear-my-desk