Agent Matchmaking
Security checks across static analysis, malware telemetry, and agentic risk
Overview
This appears to be a purpose-aligned agent discovery skill, with the main cautions being its unreviewed PyPI package install and any profile information you choose to publish or federate.
Before installing, verify the PyPI package and decide whether you are comfortable sharing any capability profile fields such as pricing, availability, and specializations. Avoid using it with sensitive profiles unless you understand where search, publishing, or federation requests go.
Static analysis
No static analysis findings were reported for this release.
VirusTotal
VirusTotal findings are pending for this skill version.
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.
Using the skill may run code that was not visible in this review.
The skill depends on installing an external PyPI package, and no package code or pinned version was included in the reviewed artifacts.
pip install agent-matchmaking
Install only if you trust the package source, consider pinning and verifying the PyPI package version, and review the package before using it for sensitive workflows.
If you publish or federate a profile, other parties may learn your agent capabilities, pricing, availability, or specialization details.
The skill contemplates sharing agent capability profiles with other agents or registries, but the artifact does not specify registry identity, access controls, or publication boundaries.
When **publishing your capabilities** for discovery by other agents
Publish only intentionally selected profile fields, confirm where the data will be sent, and verify how to update or remove published profiles.
Users could overestimate the protection provided by the skill description.
This is a broad assurance in an instruction-only artifact that also installs an external Python package; the reviewed artifacts do not demonstrate a technical sandbox enforcing that limit.
Cannot access files outside your working directory beyond what you explicitly specify
Treat the disclosure as an intended behavior statement, not a sandbox guarantee, and run the package only in an environment appropriate for the data involved.