ETH24

PassAudited by VirusTotal on May 12, 2026.

Overview

Type: OpenClaw Skill Name: eth24 Version: 0.1.0 The OpenClaw AgentSkills skill bundle 'eth24' is designed to crawl social media and RSS feeds, rank content using AI, and publish daily digests. All network requests (X API, Grok AI, Anthropic AI, Typefully API) and file system operations (reading config, writing output JSON/text/images) are directly aligned with this stated purpose. Environment variables for API keys are handled appropriately. The `SKILL.md` and `commands/eth24.md` instructions for the AI agent are functional and do not contain any malicious prompt injection attempts. There is no evidence of data exfiltration to unauthorized endpoints, persistence mechanisms, or other intentional harmful behaviors.

Findings (0)

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 malicious or spammy tweet/RSS item could try to steer the AI ranking or wording, leading to a misleading digest or draft.

Why it was flagged

The model prompt includes raw tweet and RSS content, which may contain untrusted text. This is central to ranking a social-media digest, but the generated JSON can be influenced by the input content.

Skill content
RAW DATA:\n{context[:12000]}
Recommendation

Review ranked.json, cli.txt, or thread.txt before publishing or relying on the digest.

What this means

Running tweet mode with Typefully credentials can create a draft in the connected social publishing account.

Why it was flagged

In tweet mode, the skill can send generated digest text to Typefully to create a draft. This is disclosed and purpose-aligned, but it is an account-modifying API action.

Skill content
if social_set_id and typefully_key:\n            posts = [{"text": text}]\n            create_draft(social_set_id, posts)
Recommendation

Use CLI mode for local previews, keep Typefully credentials scoped, and review drafts before publishing.

What this means

The skill can spend API quota and access the connected X/xAI/Anthropic/Typefully services according to the tokens provided.

Why it was flagged

The skill uses provider credentials for the expected crawl and ranking integrations. The registry metadata says no required env vars, so users should not miss these credential requirements.

Skill content
XAI_API_KEY=...           # xAI API key\nX_BEARER_TOKEN=...        # X API v2 bearer token\nANTHROPIC_API_KEY=...     # Anthropic API key
Recommendation

Use least-privilege, revocable API keys and monitor provider usage/costs.

What this means

Dependency behavior could change over time if installed without a lockfile or pinned versions.

Why it was flagged

The Python dependencies are not version-pinned. This is common for simple projects, but it means future installs may resolve different package versions.

Skill content
feedparser\nPillow\nhttpx
Recommendation

Install in a virtual environment and consider pinning dependency versions before routine use.