skills/smarzban/tsk/tsk-coordinator/Gen Agent Trust Hub

tsk-coordinator

Pass

Audited by Gen Agent Trust Hub on Nov 11, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill relies on shell command execution via specific CLI tools to orchestrate agents and manage the filesystem.
  • It utilizes rex for session and window management (e.g., rex window new, rex split).
  • It utilizes tsk for task board management (e.g., tsk list, tsk reply, tsk edit, tsk clean).
  • It performs repository operations using git (e.g., git worktree remove, git branch -D).
  • [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted data from external task trackers to construct briefs for other agents.
  • Ingestion points: Reads task notes, issue bodies, and tracker metadata from the tsk board or third-party trackers.
  • Boundary markers: Not explicitly defined in the task brief template, but utilizes a "Tasks" section in HANDOFF.md to scope work.
  • Capability inventory: The coordinator has capabilities for file system modification (worktrees), command execution (rex, tsk), and credential setup/merging if granted by the user.
  • Sanitization: Implements a dedicated "Review gate" process where an independent reviewer agent scans for security boundaries, credential leaks, and public writes. Includes a mandatory "Live test" phase that requires evidence (logs, screenshots) to verify outcomes against the actual environment.
  • [EXTERNAL_DOWNLOADS]: References external documentation from the tool vendor's website.
  • Points to gettsk.sh for documentation on project settings and storage configurations.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 11, 2026, 02:33 AM