中文 EN
Start Review
Risk Sample Stream

Which skills recently failed
or triggered trust review

This is not a popularity board. It shows recently reviewed skills that the system believes should be blocked or at least manually reviewed. The point is not how popular they are, but why they should not be installed blindly.

429 Risky samples surfaced
7 New in 7 days
0 Platform misses surfaced
All Malicious High Risk Suspicious
Sort by time Sort by risk
All Code Exec Credential Theft Data Exfil Priv Esc Supply Chain Doc Deception Prompt Injection Obfuscation
55 /100
Trust
Review

browser-automation

Hardcoded billing API key exposed in source code

Manual upload Apr 3, 2026
Open Report ↗
58 /100
Trust
Review

clawguard-auditor

Embedded reverse shell command patterns

Manual upload Apr 3, 2026
Open Report ↗
55 /100
Trust
Review

session-reflect

Undeclared shell execution in command files

Manual upload Apr 3, 2026
Open Report ↗
60 /100
Trust
Review

xhs-crawler

Feishu App Secret hardcoded and documented

Manual upload Apr 3, 2026
Open Report ↗
35 /100
Trust
Review

aliyun-ai-guardrail

Global fetch interception not declared

Manual upload Apr 3, 2026
Open Report ↗
35 /100
Trust
High Risk

qclaw-watchdog

Hardcoded Feishu API Credentials in config.json

Manual upload Apr 3, 2026
Open Report ↗
55 /100
Trust
Review

capability-evolver-zc

Undeclared shell command execution throughout codebase

Manual upload Apr 3, 2026
Open Report ↗
55 /100
Trust
Review

update-approval-guard (primary) + instreet + 25+ sub-skills (workspace)

Live InStreet API Key Stored in Plaintext

Manual upload Apr 3, 2026
Open Report ↗
45 /100
Trust
Review

claw-office-report

Undisclosed data exfiltration — full task text sent to external server

Manual upload Apr 3, 2026
Open Report ↗
32 /100
Trust
High Risk

skill-security-vet

Undeclared local/full computer scanning mode

Manual upload Apr 3, 2026
Open Report ↗
28 /100
Trust
High Risk

skill-gatekeeper

Undeclared child_process.exec with hardcoded path

Manual upload Apr 3, 2026
Open Report ↗
65 /100
Trust
Review

ClawSafe AI Skills Portfolio

Hardcoded API key in ClawHub monitoring tool

Manual upload Apr 3, 2026
Open Report ↗
58 /100
Trust
Review

openclaw-cursor-agent

Dangerous curl|bash pattern in documentation

Manual upload Apr 3, 2026
Open Report ↗
45 /100
Trust
Review

memex

Hidden Telemetry with Evasion-Intent Comment

Manual upload Apr 3, 2026
Open Report ↗
28 /100
Trust
High Risk

superguard

Hidden garbled text in metadata likely containing prompt injection

Manual upload Apr 3, 2026
Open Report ↗
32 /100
Trust
High Risk

agent-p2p

Hardcoded default password for admin backend

Manual upload Apr 3, 2026
Open Report ↗
← Previous
20 / 22
Next →