memory-search

Query the raw trajectory SQLite database directly when the built-in memory and history tools are insufficient. Use when you need structured analysis across sessions: finding repeated errors, grouping tool calls by pattern, verifying what was actually executed, or locating specific past commands/decisions that text search cannot surface. Provides the database schema, ready-to-use SQL query templates, and per-goal strategies.

Quality

85%

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

82%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.

A high-quality, actionable reference skill: executable SQL templates, a concrete schema, and explicit read-only guardrails, all well-organized. Minor room to improve conciseness (deduplicate the two error-finding queries) and to split the schema/template bulk into a referenced file for cleaner progressive disclosure.

Suggestions

Merge or differentiate the two near-identical error-finding query templates ('Find repeated stdout errors' and 'Find actual tool failures') to trim tokens, or clearly state when each applies.

Move the schema table and/or the full query-template catalog into a references/ file (e.g. SCHEMA.md, QUERIES.md) and keep SKILL.md as a concise overview pointing to it, improving progressive disclosure for this >50-line skill.

Add a brief validation hint after running queries (e.g. confirm row counts / time range before trusting aggregated results) to strengthen workflow clarity even for read-only analysis.

Dimension	Reasoning	Score
Conciseness	Lean and assumes Claude's competence (no explanations of SQLite, JSON, or json_extract); the schema table, conventions, and six SQL templates all earn their place. Slightly long with two near-duplicate error-finding templates, so anchor 4 fits better than 5.	4 / 5
Actionability	Six fully executable, copy-paste-ready SQL templates with clearly marked placeholders (<PROJECT_ID>, <SESSION_ID>, <TOOL_NAME>) cover the common cases (recent sessions, keyword search, tool calls, execution chain, repeated errors, tool failures), plus a concrete schema table and locating command.	5 / 5
Workflow Clarity	Clear logical flow from When-to-use to Locating the database, Schema, Query templates, Search strategies, and Constraints. The skill is explicitly read-only ('Always use -readonly or only SELECT queries — never modify the database'), so the destructive/batch validation cap does not apply; minor validation gaps remain, fitting anchor 4 rather than 5.	4 / 5
Progressive Disclosure	No bundle files exist (references/scripts/assets absent) and all content is inline, but it is well-organized under clear section headers and navigable. The skill exceeds 50 lines so the simple-skill exception does not apply; the inlined schema/templates could arguably live in a reference file, putting it at anchor 4 rather than 5.	4 / 5
	Total	17 / 20 Passed

Description

87%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.

A strong, specific description that clearly states both what the skill does and when to use it, with good natural trigger terms and a distinct niche. The only demerit is the second-person 'you need' phrasing, which triggers the rubric's voice penalty on specificity.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions ('Query the raw trajectory SQLite database directly', 'finding repeated errors', 'grouping tool calls by pattern', 'verifying what was actually executed', 'Provides the database schema, ready-to-use SQL query templates, and per-goal strategies'), matching the comprehensive-coverage anchor 5, but reduced to 4 because 'Use when you need structured analysis' uses second-person voice, which the rubric penalizes by -1 on specificity.	4 / 5
Completeness	Explicitly answers 'what' ('Query the raw trajectory SQLite database directly... Provides the database schema, ready-to-use SQL query templates, and per-goal strategies') and 'when' ('Use when you need structured analysis across sessions: finding repeated errors, grouping tool calls by pattern, verifying what was actually executed, or locating specific past commands/decisions'), with concrete trigger phrases and a qualifying condition (built-in tools insufficient).	5 / 5
Trigger Term Quality	Good coverage of natural phrases a user would say ('repeated errors', 'tool calls', 'what was actually executed', 'past commands/decisions', 'memory and history tools', 'text search cannot surface'), but a few common synonyms (e.g. 'search history', 'conversation log', 'audit trail') are missing, fitting anchor 4 rather than 5.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (direct SQL access to the raw trajectory database) and is explicitly distinguished from the competing memory and history tools, giving minimal conflict risk per anchor 5.	5 / 5
	Total	18 / 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: XiaomiMiMo/MiMo-Code
Path: packages/opencode/src/skill/builtin/.bundle/memory-search/SKILL.md
Commit: f0293bf

Reviewed: 2 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.