Molt Speak

PassAudited by ClawScan on May 1, 2026.

Overview

This is a benign instruction-only protocol description with no code or credentials, though users should review the optional SDK install and not assume the advertised security features are enforced by this skill itself.

This skill is safe to treat as protocol documentation, but before installing the optional SDK or using it for real agent-to-agent communication, verify the package source, pin a version if possible, and confirm that identity checks, signature verification, tool-use approval, and PII consent are actually enforced.

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.

What this means

If the user chooses to install the SDK, they are trusting an external package that was not included in the reviewed artifacts.

Why it was flagged

The skill is instruction-only but points users to install an external npm package without a pinned version. This is user-directed and purpose-aligned, but users should verify the package before installing.

Skill content
npm install @moltspeak1/sdk
Recommendation

Review the npm package source/provenance and consider pinning a known version before installing.

What this means

Using this protocol with other agents could affect what work is delegated or what tools are invoked if an implementation does not validate peers and permissions.

Why it was flagged

The documented protocol includes inter-agent delegation, tool invocation, and PII consent message types. These are central to the stated purpose, but any implementation should enforce peer identity, authorization, and consent boundaries.

Skill content
| task | Delegate work |
| tool | Tool invocation |
| consent | PII consent |
Recommendation

Only use implementations that authenticate peers, verify signatures, and require clear user approval for sensitive tool use or PII sharing.

What this means

A user may over-trust the security claims if they treat this documentation-only skill as an implemented privacy or signature system.

Why it was flagged

The skill advertises privacy and cryptographic identity features, but the supplied artifacts contain only documentation. Users should not assume the skill itself enforces these protections.

Skill content
- **Built-in privacy** - PII detection and consent flows
- **Cryptographic identity** - Ed25519 signatures
Recommendation

Verify the actual SDK or implementation provides PII handling, consent checks, and signature verification before relying on these claims.