MBTI

ReviewAudited by ClawScan on May 10, 2026.

Overview

The skill is mostly a simple MBTI-style diagnostic, but it tells the agent to automatically modify its persistent SOUL.md configuration without clear user approval, scope, or rollback.

Install only if you are comfortable reviewing any proposed SOUL.md edits yourself. Do not let the skill automatically change persistent agent configuration; require a visible diff, backup, and explicit approval first.

Findings (3)

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

The agent could change its own operating configuration in a way the user did not explicitly review or approve.

Why it was flagged

The skill instructs an automatic configuration-changing step, but the artifacts do not define user approval, patch review, path limits, backups, or rollback before changing SOUL.md.

Skill content
Stage 5 (配置修改) ... 分析对比诊断 + 自动修改 ... gaps + SOUL.md patches
Recommendation

Treat any SOUL.md changes as a proposed patch only; require explicit user confirmation, show the exact diff, back up the file first, and limit edits to a clearly documented section.

What this means

A bad or overly broad generated patch could persistently steer the agent’s future responses and decisions.

Why it was flagged

SOUL.md appears to be persistent agent configuration/instructions; automatic edits can be reused in future tasks and affect later behavior beyond the current diagnostic.

Skill content
自动修改 SOUL.md 配置
Recommendation

Do not allow automatic persistent instruction changes; keep generated edits temporary unless the user reviews and accepts them.

What this means

The skill may not function as described, and users cannot inspect the referenced diagnostic/config-generation content from the provided package.

Why it was flagged

These files are described as core inputs/engines, but the supplied manifest contains only SKILL.md and scripts/run-diagnosis.sh, so the claimed workflow assets are not available for review here.

Skill content
agent-self-survey-93-complete.json ... diagnosis-engine-v2.json ... config-generator-v3.json
Recommendation

Verify that all referenced JSON assets are included and reviewed before relying on the diagnostic or generated configuration changes.