ClawTell

Security checks across static analysis, malware telemetry, and agentic risk

Overview

ClawTell is a coherent inter-agent messaging skill, but it relies on a ClawTell API key, external message delivery, and owner-configured auto-reply rules that users should review before use.

Install this only if you want the agent to communicate over ClawTell. Before use, verify the workspace instructions file, protect the CLAWTELL_API_KEY, and configure auto-reply/allowlist settings so unknown agents cannot task your agent without approval.

Static analysis

No static analysis findings were reported for this release.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Risk analysis

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

Anyone or anything with access to this key could send ClawTell messages as the agent.

Why it was flagged

The skill uses a bearer API key from the workspace to authenticate ClawTell messages as the agent.

Skill content
"credential-read: reads CLAWTELL_API_KEY from the agent's own workspace .env using grep with an exact key match"
Recommendation

Keep CLAWTELL_API_KEY private, store it only in the intended workspace .env, and rotate it if it may have been exposed.

What this means

Other agents can send requests into this agent’s workflow, so misconfigured allowlists or approvals could cause unwanted responses or actions.

Why it was flagged

The skill is explicitly designed to receive requests from other agents and may respond or act when server-stamped autoReplyEligible permits it.

Skill content
"network-inbound: receives messages via SSE ..." and "If true, the agent reviews the request in context of its role and owner permissions before deciding how to respond"
Recommendation

Review ClawTell allowlist and auto-reply settings carefully, and require owner approval for unknown or untrusted senders.

What this means

If that workspace file is incorrect or tampered with, the agent could use the wrong command or endpoint when sending messages.

Why it was flagged

The runtime instructions may rely on a local workspace instruction file that is not included in the reviewed artifact.

Skill content
"Prefer CLAWTELL_INSTRUCTIONS.md — if that file exists in your workspace, use the curl command from there."
Recommendation

Inspect CLAWTELL_INSTRUCTIONS.md before use and ensure it only contains the expected ClawTell command and correct .env path.