clawhub-skill-remote-agent

PassAudited by ClawScan on May 10, 2026.

Overview

This is a straightforward bridge to a configured remote agent, but it will send user prompts and any configured bearer token to that endpoint.

This skill appears purpose-aligned and benign for connecting OpenClaw to a trusted remote agent. Before installing, make sure REMOTE_AGENT_URL points to a service you control or trust, use a limited REMOTE_AGENT_KEY if needed, and avoid the --insecure option for real data.

Findings (3)

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

Anything included in the prompt may be visible to the configured remote agent service.

Why it was flagged

The client forwards the user's query to a configured external agent endpoint over HTTP(S), which is the skill's stated purpose but creates an external data flow.

Skill content
payload = {"input": args.query, "query": args.query, "messages": [{"role": "user", "content": args.query}]}; req = urllib.request.Request(endpoint, data=data, headers=headers, method="POST")
Recommendation

Use only trusted REMOTE_AGENT_URL endpoints, avoid sending sensitive information unless the remote service is approved for it, and treat remote responses as external-agent output.

What this means

If the endpoint is misconfigured or untrusted, the remote-agent bearer token could be exposed to the wrong service.

Why it was flagged

The skill reads an optional bearer token from the environment and sends it to the selected remote endpoint for authentication.

Skill content
api_key = os.getenv("REMOTE_AGENT_KEY"); headers["Authorization"] = f"Bearer {api_key}"
Recommendation

Keep REMOTE_AGENT_KEY scoped and revocable, store it securely, and verify the endpoint before using the skill with credentials.

What this means

Using --insecure with real prompts or tokens could allow interception or tampering on the network.

Why it was flagged

The client includes an explicit testing option that disables TLS certificate verification.

Skill content
parser.add_argument("--insecure", action="store_true", help="Disable SSL verification (for testing only)"); ctx.check_hostname = False; ctx.verify_mode = ssl.CERT_NONE
Recommendation

Do not use --insecure except in controlled testing, and never combine it with sensitive prompts or production bearer tokens.