task-progress

Pass

Audited by Gen Agent Trust Hub on Apr 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands such as mc mirror and copaw-sync for synchronizing task data.
  • [DATA_EXFILTRATION]: The skill transmits local task progress logs and history to a remote storage location defined by the ${HICLAW_STORAGE_PREFIX} environment variable. This is consistent with the skill's stated purpose of progress tracking and state persistence.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection attack surface as it reads external data from files like spec.md, plan.md, and historical progress logs.
  • Ingestion points: Reads from ~/.copaw-worker/<your-name>/shared/tasks/{task-id}/spec.md, plan.md, and progress/*.md.
  • Boundary markers: None specified in the instructions for reading these files.
  • Capability inventory: Executes shell commands (mc mirror, copaw-sync) and performs file system writes.
  • Sanitization: No explicit sanitization or validation of the content read from the task files is mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 6, 2026, 08:12 AM