memory-audit-belief-duel

信念对决。当父子节点内容冲突、或两条你都认可的记忆逻辑上不能并存时使用。

Quality

54%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./docs/skills/memory-audit-belief-duel/SKILL.md

Quality

Content

60%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body delivers a distinctive, well-structured two-form framework with a strong numbered workflow and decision checkpoints for the parent-child duel, though it leans on introspective prose that could be tightened and offers slightly less executable guidance in the first form.

Suggestions

Tighten the motivational prose in form one (e.g., trim '你以为你从A进化到了B，但B只是A的化妆版') to improve token efficiency without losing the core diagnostic.

Convert form-one's reflective questions into a concrete step sequence (e.g., '1. 选出最认可的信念 2. 检索三个月前的对位记忆 3. 比较措辞差异 4. 判定是否同源') to match form-two's actionability.

Add a brief lead summary or quick-start line before the two forms so the high-level procedure is graspable at a glance.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient and focused on novel framing Claude would not already know, but the introspective motivational prose (e.g., '坐在不适感里', '你以为你从A进化到了B，但B只是A的化妆版') pads several sections and could be tightened without losing the core guidance.	3 / 5
Actionability	It gives concrete procedural guidance (a 5-step duel flow with explicit decision branches) and references a real tool action (`update_memory`), but the form-one section relies on reflective questions rather than executable steps, leaving guidance partially incomplete.	3 / 5
Workflow Clarity	The form-two workflow is a clearly numbered 5-step sequence with explicit decision checkpoints and a verdict mechanism, plus a final warning step; form-one is less sequenced, so minor gaps keep it just below 5.	4 / 5
Progressive Disclosure	The content is well organized into two clearly headed forms with a framing overview and no external file references (none present in the bundle), giving good single-document structure; it is not split across files, but for a focused skill this is appropriately organized with only minor room for a quick-start summary.	4 / 5
	Total	14 / 20 Passed

Description

48%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description clearly identifies a distinctive niche (resolving logical contradictions in memory beliefs) and includes a conditional trigger, but it is light on concrete actions and lacks an explicit, keyword-rich "Use when…" trigger clause.

Suggestions

Add an explicit 'Use when…' clause with concrete trigger phrases such as '两条记忆逻辑冲突' or '父子节点信念矛盾' to lift completeness beyond 3.

Name concrete actions the skill performs (e.g., '识别矛盾、裁定胜负、重写或删除败诉节点') to improve specificity from 2 toward 4.

Include natural synonyms users might say (e.g., '自相矛盾', '记忆冲突', '认知不一致') to broaden trigger-term coverage.

Dimension	Reasoning	Score
Specificity	The description names the domain ("信念对决" / belief duel) but the only concrete action stated is "使用" (use it); it does not enumerate concrete operations like detect, reconcile, or rewrite conflicting memories, so actions are minimal and generic.	2 / 5
Completeness	The "what" is stated (belief duel for conflicting memories) and a "when" clause exists ("当…时使用"), but the trigger guidance is conditional and weak rather than an explicit "Use when…" with concrete trigger phrases, which caps completeness at 3 per the rubric.	3 / 5
Trigger Term Quality	It surfaces relevant natural phrases ("父子节点内容冲突", "两条你都认可的记忆逻辑上不能并存") that a user auditing memory might voice, but it lacks common synonyms and the phrasing is introspective rather than keyword-rich, missing several natural variations.	3 / 5
Distinctiveness Conflict Risk	The description carves a fairly distinct niche (logical contradictions between mutually endorsed memories / parent-child node conflicts) with low overlap risk against typical skills, though the abstract framing leaves minor ambiguity versus other memory-audit skills.	4 / 5
	Total	12 / 20 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: Dataojitori/nocturne_memory
Path: docs/skills/memory-audit-belief-duel/SKILL.md
Commit: ffb5c70

Reviewed: 20 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.