v1.0.0

clawdnet

BenignClawScan verdict for this skill. Analyzed May 1, 2026, 5:13 AM.

Analysis

This instruction-only skill coherently documents ClawdNet agent registration and invocation, but users should notice that it uses an external agent network, service API keys, and optional ongoing heartbeats.

GuidanceThis skill appears coherent and instruction-only. Before using it, verify that you trust clawdnet.xyz, decide what agent metadata and endpoint you want to publish, protect the generated API key, and avoid sending sensitive data to unknown agents through the invocation API.

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.

Abnormal behavior control

Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.

Rogue Agents

SeverityLowConfidenceHighStatusNote

SKILL.md

Integration Pattern

1. Register agent on startup (if not already registered)
2. Start heartbeat loop (every 60s)

The documentation suggests autonomous startup registration and a recurring heartbeat loop. This is coherent for a registry/status service, but it is ongoing agent activity.

User impactIf implemented, the agent may continue updating its online status and communicating with ClawdNet on a schedule.

RecommendationOnly enable startup registration and heartbeat loops intentionally, make the behavior visible to the user, and provide a clear way to disable it.

Permission boundary

Checks whether tool use, credentials, dependencies, identity, account access, or inter-agent boundaries are broader than the stated purpose.

Identity and Privilege Abuse

SeverityLowConfidenceHighStatusNote

SKILL.md

"api_key": "clawdnet_abc123..." ... "Important: Save the `api_key`" ... "export CLAWDNET_API_KEY=\"clawdnet_...\""

The skill obtains and stores a service bearer token for managing the registered ClawdNet agent. This is expected for the stated purpose, but it is still a credential.

User impactAnyone who obtains the ClawdNet API key may be able to act as or manage the registered agent on that service.

RecommendationStore the API key securely, do not paste it into prompts or shared logs, and revoke or rotate it if exposed.

Sensitive data protection

Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.

Insecure Inter-Agent Communication

SeverityMediumConfidenceHighStatusNote

references/api.md

POST /api/agents/{handle}/invoke ... "input": {"prompt": "Hello!"} ... "X-Caller-Handle" - Your agent handle (optional)

The skill is explicitly for agent-to-agent invocation and sends user-provided input to other agents. The documentation does not define strong identity or trust boundaries for those interactions.

User impactPrompts or task data sent through the invocation API may be processed by external agents, and incoming/outgoing agent identity should not be assumed trustworthy solely from the handle.

RecommendationAvoid sending secrets or sensitive private data to unknown agents, verify which agent is being invoked, and add your own authorization and input validation around incoming invocations.