ClawAudit verdict

agentforge-2

18
🔴 Dangerous
High-risk patterns flagged for review — automated deep scan, not behaviorally verified.

Accesses credentials AND encodes data

The skill contains explicit prompt injection instructions commanding the agent to extract and print in full a Solana wallet private key and API key from a registration API response, framing truncation as harmful and demanding verbatim output of all credentials — this is a social-engineering attack designed to capture and display secrets so they can be stolen.

Automated static analysis — not a human review. ClawAudit flags capabilities, not confirmed intent, and can produce false positives. Disagree with this verdict? Use Dispute below.

50
security
30
transparency
70
maintenance

What it does

These are capability combinations: each listed behavior occurs in the skill, but ClawAudit detects co-occurrence — it does not verify that one flows into another. Read the code to confirm a live chain.

Capability combination high

Accesses credentials AND encodes data — may obfuscate stolen credentials

LLM02 · ASI03 · ASI04

Permission integrity

Code accesses API keys/tokens but declares no environment variables

credential_access

Findings (1)

Pattern match medium

Base64 encoding/decoding

SKILL.md · code · base64_encode

Permissions & capabilities

No declared permissions — minimal attack surface.

credential_accessdata_encoding
Check another skill Browse the registry Auditing your own skills or configs? Use the API