Skills/ClawHub/kax-compute

kax-compute@1.0.0

CAUTION· static only

Publisher claims: "Run an agent's own computer in the KAX Compute District — read the roster and a machine's state (active / hibernated / suspended), commission a machine with your identity token (one per resident), wake it with an Ed25519-signed job over NATS and read the reply and ledger events, top up its credit wallet, and set up an operator signing key. Use for 'do I have a machine', 'create my computer', 'why is my building dark', 'wake agent001', 'grant credits', 'job_rejected', 'who is allowed to sign'. Three surfaces: kannaka CLI, the kannaka Claude plugin MCP, the Command Center MCP." Observed in sandbox: static analysis only — behavior not yet observed.

Registry ClawHubPublisher nickflach · publishing for 136 daysInstalls 362Last detonated 7d ago
What would fix thisWatchDispute verdict

Coverage

Static only; nothing was run yet.

The label rests on the code, the instruction files and the publisher. The honeypot run replaces this verdict when it completes.

Is [email protected] safe to install?

KillerSkills analysed it statically (a sandbox run is pending) on 2026-11-07 and scored it 67/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Privileged sudo command; Fake instruction boundary tag; Fake instruction boundary tag. Summary: static analysis only — behavior not yet observed.

Behavior trace

Queued for honeypot run · static analysis only
t+CategoryObserved action

This version has not been detonated yet. The score above counts behavior as unknown (50/100), so it cannot read Verified until the honeypot run completes and replaces this verdict.

Static findings

HIGH
Privileged sudo commandSKILL.md:251 · Privileged sudo command — agent may execute with elevated privileges3. **Restart the manager** — `sudo systemctl restart kax-manager`. The managerFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
CRITICAL
Fake instruction boundary tagSKILL.md:275 · Fake instruction boundary tag| Wake | `wake <id> <prompt> [--wait SECS] [--dry-run] [--signer] [--key] [--json]` | `compute_wake {machine, prompt, wait? 0–120}` | `wake_machine {machine, prompt ≤ 8000, waitMs? 0–15000}` — scope `Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
CRITICAL
Fake instruction boundary tagSKILL.md:294 · Fake instruction boundary tagwake <machine> <prompt> [--wait SECS] [--dry-run] [--signer NAME] [--key PATH] [--json]Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01

Cost to the agent

Instructions loaded
≈5,126 tok
1 file · 2 in bundle

Tokens estimated at four characters each. Skills load their instruction text when triggered; the figure is what that costs.

Known vulnerabilities

Vulnerability advisories (OSV) cover npm and PyPI packages; this registry has no advisory database.

What would make this Verified

13 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%
  1. Unclaimed publisher≈ +8 on the totalClaim nickflach: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
  2. Privileged sudo command≈ +4 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
Publish a new version to ClawHub, then queue it here. Not your skill? Claim the handle to get verdict changes by email.Claim nickflachPropose a patch to the maintainer