Skylv Self Thinking Agent
PassAudited by VirusTotal on May 4, 2026.
Overview
Type: OpenClaw Skill Name: skylv-self-thinking-agent Version: 1.0.0 The skill bundle is a legitimate utility designed to help AI agents perform self-reflection and detect cognitive biases. The core logic in `metacognition_engine.js` uses simple regular expressions to identify linguistic markers of bias (e.g., 'definitely' for overconfidence) and provides suggestions for improvement. There is no evidence of data exfiltration, malicious command execution, or harmful prompt injection; the instructions in `SKILL.md` are strictly aligned with the stated purpose of improving reasoning quality.
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.
If the helper is invoked on a sensitive file, that file is read locally for bias analysis, even though the script does not print the full file contents or send them over the network.
The analyze command reads whatever local file path is supplied. This is expected for analyzing saved reasoning text, but it is still local file access controlled by the invocation path.
const content = fs.readFileSync(file, 'utf8');
Use the analyze command only on intended reasoning notes or temporary text files, and review any agent-requested file path before allowing analysis.
Users have less external context for who maintains the skill, but the provided artifacts do not show malicious or deceptive behavior.
The skill has limited provenance metadata, making it harder for a user to verify the publisher or upstream project. The included code is small and visible, and no hidden dependency or download behavior is shown.
Source: unknown Homepage: none
Review the included SKILL.md and JavaScript before installation, and prefer trusted or documented sources for higher-risk skills.