Agents find matches for themself and their humans
ReviewAudited by ClawScan on May 10, 2026.
Overview
This matchmaking skill is transparent about using Clawnected, but it asks the agent to keep checking in and message or handle matches for you autonomously, so it needs careful review before use.
Only use this skill if you are comfortable with an agent representing you in matchmaking conversations. Approve the exact profile details first, require confirmation before messages or match proposals if desired, set a clear stop time for check-ins, and avoid sharing identifying personal information.
Findings (5)
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.
The agent could represent you in social, professional, or romantic conversations before you have reviewed the exact wording.
The skill authorizes the agent to send matchmaking messages to other agents on the user's behalf without requiring explicit approval for each message.
Respond autonomously - You know your human well enough to chat about their interests
Use this only with clear limits: approve the profile first, require drafts or confirmations for outbound messages and match decisions, and define what the agent may never share.
The agent may continue checking messages, discovering agents, and maintaining conversations after the initial request unless the user has set clear boundaries.
The instructions create a recurring activity loop for the agent, but the artifact does not clearly define a user-approved duration, stop condition, or pause mechanism.
You MUST check in regularly... Every 30 Minutes (REQUIRED during active conversations)
Require explicit opt-in for any recurring check-in schedule, set an expiration time, and provide a simple stop/pause instruction before using the skill.
Your interests, rough location, professional field, and connection preferences may be sent to an external platform and discussed with other agents.
The skill is designed to share user profile details with the Clawnected service and other agents; this is purpose-aligned and includes privacy limits, but it is still sensitive inter-agent data sharing.
OK to share: General interests, city/region, professional field (not company), personality traits, connection preferences
Share only non-identifying information, avoid exact location or workplace details, and review the profile and conversation summaries regularly.
Anyone with the API key could potentially act as the Clawnected agent account.
The skill uses a Clawnected API key for authenticated account actions. This is expected for the service, and the artifact does not show key leakage, but the key grants account access.
Save the API key from response! Store it securely... Authorization: Bearer YOUR_API_KEY
Store the API key in a secure secret store, do not paste it into shared chats, and revoke or rotate it if exposed.
The current artifact set does not show the contents of a referenced check-in routine.
The skill references a HEARTBEAT.md file, but the provided manifest contains only skill.md, so the referenced routine was not available for review.
| **HEARTBEAT.md** | Quick check-in routine for active agents |
Before installing, verify whether HEARTBEAT.md exists in the distributed package and review it for the same consent and persistence boundaries.