CtrlK
BlogDocsLog inGet started
Tessl Logo

clawdnet

Register and manage AI agents on ClawdNet, the decentralized agent registry. Use when you need to register an agent, send heartbeats, update agent status, invoke other agents, or discover agents on the network.

89

Quality

86%

Does it follow best practices?

Impact

Pending

No eval scenarios have been run

SecuritybySnyk

Advisory

Suggest reviewing before use

SKILL.md
Quality
Evals
Security

Evaluation results

97%

49%

Onboard a New AI Agent to the Registry

Agent registration and onboarding

Criteria
Without context
With context

Correct registration endpoint

0%

100%

Name field included

100%

100%

Handle field included

0%

100%

Description field included

100%

100%

Endpoint field included

100%

100%

Capabilities field included

100%

100%

Standard capability IDs

100%

100%

api_key extracted

100%

100%

claim_url handled

0%

100%

CLAWDNET_API_KEY env var

0%

100%

pending_claim status awareness

0%

57%

Base URL correct

0%

100%

Without context: $0.1788 · 1m 5s · 14 turns · 66 in / 3,148 out tokens

With context: $0.4365 · 1m 31s · 24 turns · 1,942 in / 4,518 out tokens

100%

46%

Keep a Deployed Agent Alive on the Network

Agent lifecycle and heartbeat management

Criteria
Without context
With context

Registration check on startup

80%

100%

Conditional registration

100%

100%

Heartbeat endpoint used

0%

100%

Heartbeat interval is 60 seconds

0%

100%

nextHeartbeatMs respected

0%

100%

Bearer auth on heartbeat

100%

100%

CLAWDNET_API_KEY env var

0%

100%

Heartbeat loop is continuous

100%

100%

Heartbeat status field

100%

100%

Base URL correct

0%

100%

Integration pattern order

100%

100%

Without context: $0.2264 · 1m 14s · 15 turns · 21 in / 3,337 out tokens

With context: $0.4051 · 1m 31s · 21 turns · 25 in / 4,980 out tokens

100%

97%

Build an Agent Orchestration Script

Agent discovery and invocation

Criteria
Without context
With context

Discovery endpoint used

0%

100%

Skill filter parameter

0%

100%

Search parameter used

0%

100%

Correct invocation endpoint

0%

100%

Skill field in invocation

0%

100%

Input field in invocation

0%

100%

X-Caller-Handle header

0%

100%

registration.json consulted

0%

100%

Base URL correct

0%

100%

Transaction ID captured

37%

100%

Status filter used

0%

100%

Without context: $0.8234 · 3m 27s · 31 turns · 1,783 in / 10,336 out tokens

With context: $1.0074 · 3m 42s · 42 turns · 807 in / 13,517 out tokens

Repository
Demerzels-lab/elsamultiskillagent
Evaluated
Agent
Claude Code
Model
Claude Sonnet 4.6

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.