Skillv1.0.0

ClawScan security

Pilot Sales Pipeline Setup · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

BenignApr 23, 2026, 4:43 AM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: The skill's instructions, required tools, and actions are coherent with its stated purpose (deploying a 4-agent sales pipeline); it is instruction-only and does not request unrelated credentials or perform hidden installs.
Guidance: This skill is internally consistent for deploying a 4-agent sales pipeline, but take these precautions before running it: 1) Ensure pilotctl and clawhub are legitimate, up-to-date binaries from vendors you trust. 2) Review each pilot-* component that the skill instructs you to install (those components may request API keys or network access). 3) Be deliberate about the handshake/trust steps: exchanging handshakes will open agent-to-agent communication — verify peer identities and network boundaries. 4) The setup writes config to ~/.pilot/setups; inspect that file for any sensitive data before sharing. 5) Consider testing in an isolated environment (staging or VM) first, and review the project/homepage and licensing (AGPL) to confirm it meets your operational and compliance requirements.
Findings: [no-findings] expected: The regex-based scanner found nothing to analyze because this is an instruction-only skill with no code files. That absence is expected but does not replace manual review of the pilot-* components the skill instructs you to install.

Review Dimensions

Purpose & Capability: okThe name/description (deploy a 4-agent sales pipeline) lines up with the runtime instructions: it requires pilotctl and clawhub, installs pilot-* skills via clawhub, writes a local manifest, sets hostnames, and initiates peer handshakes. All required binaries and actions are proportional to the stated goal.
Instruction Scope: noteInstructions are narrowly scoped to installing pilot-* components, writing a manifest to ~/.pilot/setups, setting hostnames, and exchanging handshakes. Note: it instructs the agent/operator to install many auxiliary pilot-* skills and to open network trust relationships between agents — these are expected for a distributed pipeline but increase attack surface and should be reviewed before running.
Install Mechanism: okThis is an instruction-only skill with no install spec or external download. The only install actions are user-invoked clawhub install commands (no arbitrary URL downloads or extraction). That is lower risk for this kind of orchestration task.
Credentials: okThe skill declares no environment variables or credentials and does not attempt to read system-wide config paths beyond creating ~/.pilot/setups. This matches its stated purpose. Caveat: the pilot-* components it installs may themselves request credentials (e.g., email/CRM API keys); inspect those components before installing.
Persistence & Privilege: okalways is false and the skill does not request persistent elevated privileges or modify other skills' configurations. It writes its own manifest under the user's home (~/.pilot) which is appropriate for configuration state.