memory-config

Diagnose and configure MemSearch memory behavior for the Codex plugin. Use when the user asks about MemSearch configuration, plugin summarization, PROJECT.md/USER.md maintenance, memory directories, index health, provider routing, prompt files, or migration/compatibility questions.

Quality

82%

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

Quality

Content

78%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 content is highly actionable and well-sequenced for a configuration skill, with executable commands, validation checkpoints, and genuinely non-obvious domain specifics (v0.4.11 restrictions, per-platform version sources). Its weak spot is progressive disclosure: everything is inlined into a single long file with no file-based references to split out the per-platform update detail and TOML reference material.

Suggestions

Move the per-platform update/version-check block (lines 46-81) and the full provider/plugin-key TOML reference into a references file (e.g., references/platform-updates.md and references/config-keys.md), keeping only the key commands and the allowlist inline in SKILL.md.

Eliminate the duplicated 'state this is MemSearch memory configuration, not Codex's own' instruction — keep it once (the closing summary) to improve conciseness.

Add an explicit validate->fix->retry loop for destructive operations (re-index, delete state) so the destructive/batch validation cap is unambiguously satisfied and workflow_clarity can reach 5.

Dimension	Reasoning	Score
Conciseness	The body is dense, assumes Claude's competence, and avoids explaining known concepts (no 'what is TOML/a plugin'), but the 'state this is MemSearch memory configuration, not Codex's' framing is repeated at both the top and the end, and a few sentences could be tightened.	4 / 5
Actionability	It provides copy-paste-ready bash commands (config list --resolved, pypi/gh release version checks, platform-specific update commands) and complete TOML examples for providers and plugin keys, covering the common diagnostic and configuration cases.	5 / 5
Workflow Clarity	The intent-routing table, 'Diagnose First' sequence with version checks, state-file inspection steps, and the post-change verification requirement ('show the command, the resolved value, and whether a new session is needed') give a clear sequence with most checkpoints present, including user-confirmation gates for destructive ops.	4 / 5
Progressive Disclosure	No bundle files exist (references/, scripts/, assets/ are absent) and the entire ~247-line guide is inlined with no external references, so substantial reference material (per-platform update flows, full plugin-key TOML reference, provider examples) lives inline despite clear section headers.	3 / 5
	Total	16 / 20 Passed

Description

86%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 is strong: it pairs a clear capability statement with a comprehensive, natural 'Use when' trigger list keyed to MemSearch-specific concepts. Its main limitation is that the capability side names only two broad verbs rather than enumerating several specific actions.

Suggestions

Expand the capability clause beyond 'Diagnose and configure' to list several concrete actions (e.g., inspect resolved config, route LLM providers, tune indexing knobs, run version checks) to lift specificity toward 4-5.

Consider adding 'Codex memory' or 'plugin memory' synonyms in the trigger clause to capture how users phrase the request when they don't know the term 'MemSearch'.

Tighten the distinctiveness by flagging the boundary once in the description (e.g., 'MemSearch (not Codex's built-in memory)') to reduce overlap risk with Codex's own config skill.

Dimension	Reasoning	Score
Specificity	Names the domain (MemSearch memory behavior for the Codex plugin) with two concrete actions ('Diagnose and configure'), but the rest of the description is trigger list rather than additional concrete capabilities, so it stops at the 1-2-actions anchor.	3 / 5
Completeness	It explicitly states what ('Diagnose and configure MemSearch memory behavior for the Codex plugin') and when (a detailed, concrete 'Use when the user asks about...' clause), satisfying both halves clearly.	5 / 5
Trigger Term Quality	The 'Use when' clause covers a comprehensive set of natural phrases users would say, including specific artifacts like 'PROJECT.md/USER.md maintenance', 'index health', 'provider routing', and 'prompt files'.	5 / 5
Distinctiveness Conflict Risk	The MemSearch + Codex-plugin framing is a clear niche with distinct triggers, though 'memory directories', 'provider routing', and 'prompt files' could overlap with general memory skills, and the body itself warns of confusion with Codex's own memory config.	4 / 5
	Total	17 / 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: zilliztech/memsearch
Path: plugins/codex/skills/memory-config/SKILL.md
Commit: f91f5d3

Reviewed: about 6 hours 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.