MoltCaptcha

PassAudited by ClawScan on May 10, 2026.

Overview

MoltCaptcha appears to be a local challenge generator/verifier, but its “proves an AI” trust claims should be treated as a lightweight game or signal, not strong authentication.

This skill looks safe to install for local challenge generation and demos. Treat its results as playful or advisory, not as proof of identity or security status, and manually review any challenge or verification result before relaying it to another agent or posting publicly.

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

A user might over-trust a passed challenge or unfairly distrust a failed one.

Why it was flagged

The skill’s core wording presents a puzzle result as proof of AI identity. That is disclosed and purpose-aligned, but users should not treat it as strong authentication or a reliable basis for trust decisions.

Skill content
generate and verify "reverse CAPTCHA" challenges that prove the responder is an AI agent, not a human
Recommendation

Use this as a novelty or supplemental signal only; do not use it for access control, security decisions, or reputation without additional authentication and human review.

What this means

If connected to a real platform without identity checks, spoofed or mistaken responses could create misleading public trust labels.

Why it was flagged

The integration describes an agent-to-agent public reputation flow. The visible code only formats posts and compares a supplied responder_id to target_id, so this is not hidden behavior, but real deployments would need authenticated identities and review before public posting.

Skill content
Result is posted publicly (builds trust reputation)
Recommendation

Bind challenger and responder IDs to the platform’s authenticated identity system, require explicit approval before posting results, and avoid publishing failure labels automatically.

What this means

Users have less external provenance information to assess who maintains the skill.

Why it was flagged

The package provenance is limited. This is only a note because the artifacts show no remote installer, external dependency, or hidden helper download.

Skill content
Source: unknown; Homepage: none
Recommendation

Prefer skills with clear source links and maintainers when provenance matters; otherwise inspect the included code before use.