Axiom Distributed Science

• Risk level: Low
• Analyzed: May 1, 2026, 6:42 AM
• Findings: 1
• Version: 1.1.0

Overview

This instruction-only skill coherently documents public Axiom API queries and one optional submission endpoint, with no credentials, install code, persistence, or local data access shown.

This skill appears safe for querying public Axiom science data. Before using the suggestion endpoint, confirm you want to send the idea and any author information to axiom.heliex.net.

Findings (1)

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.

ASI02: Tool Misuse and Exploitation

Low

What this means

If used, an experiment idea and author name may be sent to the Axiom website for review.

Why it was flagged

The skill documents a curl POST that sends user-provided text and an author field to an external service. This is purpose-aligned and disclosed, but it is still an action that transmits user content outside the local environment.

Skill content

curl -X POST https://axiom.heliex.net/api/suggest.php ... -d '{"idea": "Study how network topology affects consensus speed in voter models with stubborn agents", "author": "your-name"}'

Recommendation

Only submit ideas when the user explicitly wants to do so, and avoid including private, sensitive, or unpublished information unless the user accepts that disclosure.