Muse

SuspiciousAudited by ClawScan on May 10, 2026.

Overview

Muse is coherent with its stated purpose, but it grants very broad team-history access and autonomous agent control through an authenticated external CLI without clear scope or safety boundaries.

Before installing, confirm that your team approves sharing coding sessions and knowledge-base content with Tribe/Muse, verify the npm CLI source, use a limited account, avoid `-force -all` and autonomous-agent commands unless you intend them, and monitor or stop any spawned agents.

Findings (4)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

Private code, prompts, commands, and team discussions could be searched, synced, and reused by the agent; stale or poisoned history could also influence future answers.

Why it was flagged

The skill is designed to expose broad historical coding context to the agent, but the artifacts do not define path limits, exclusions, retention, or how this context is reused across tasks.

Skill content
Give ClawBot access to your team's entire coding history... past sessions, team knowledge, and project context
Recommendation

Use only with team consent, confirm the provider's storage and retention rules, define project/path exclusions, and avoid broad sync or search unless needed.

ConcernHigh Confidence
ASI10: Rogue Agents
What this means

Autonomous agents could continue acting across the codebase beyond the user's immediate request if started without careful supervision.

Why it was flagged

The documented workflow can spawn subagents and run autonomous agent sessions on an interval, but no artifact states approval, timeout, workspace, or rollback limits.

Skill content
tribe muse spawn "Fix the login bug" fix-login ... tribe circuit auto --interval 30
Recommendation

Require explicit user approval before `muse spawn`, `circuit spawn`, or `circuit auto`; monitor active agents and use kill/stop commands promptly when finished.

What this means

ClawBot may act using the privileges of the logged-in Tribe account, including access to team data or agent controls available to that account.

Why it was flagged

The skill relies on a logged-in CLI session that may carry access to team coding history and agent-management privileges, but the artifacts do not define credential scope or least-privilege expectations.

Skill content
Requires authentication: Run `tribe login` first. Most commands need an active session.
Recommendation

Use a least-privileged account or workspace, verify what `tribe login` grants, and revoke the session when the skill is no longer needed.

What this means

The safety of the installed CLI package is outside this artifact review.

Why it was flagged

The skill depends on an external npm CLI package for its real behavior, while the submitted artifacts contain no code files for review.

Skill content
install:\n  method: npm\n  package: "@_xtribe/cli"\n  postInstall: "tribe login"
Recommendation

Install only from a trusted npm source, check the package publisher and version, and prefer pinned versions or vendor documentation before deployment.