CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

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

93%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

90%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.

An excellent, dense reference skill: executable SQL templates, precise schema documentation, and explicit safety constraints with zero padding. The only minor gaps are the absence of an explicit query-refinement feedback loop and no use of one-level-deep reference files for the detailed schema catalog.

Suggestions

Add a short feedback-loop note (e.g., 'if a query returns nothing, widen the time window or drop the session filter') to make error recovery explicit for failed or empty searches.

Consider moving the detailed part-type JSON catalog and less-common query templates to a one-level-deep references/ file, keeping SKILL.md as a leaner overview with the schema table and top templates.

DimensionReasoningScore

Conciseness

Lean and dense throughout: the schema table, part-type JSON shapes, key conventions, and constraints are all non-obvious, tool-specific facts Claude cannot know, and there is no padding or explanation of general concepts (e.g., no 'SQLite is a...' filler).

5 / 5

Actionability

Six copy-paste-ready SQL templates with clearly marked placeholders (<PROJECT_ID>, <SESSION_ID>, <TOOL_NAME>), an exact database-locate command using -readonly, exact json_extract() paths, and a per-goal strategy table covering the common analysis cases.

5 / 5

Workflow Clarity

Clear sequence (locate DB, verify schema via .tables, select template/strategy, apply constraints) with real guardrails (-readonly, LIMIT, session/time filters), but there is no explicit feedback loop for refining a query that returns nothing. The destructive/batch cap does not apply since all operations are read-only SELECTs with verification present.

4 / 5

Progressive Disclosure

Well-organized single file with clear sections, tables, and easy navigation, and nothing that clearly belongs in a separate file since the schema and templates are needed on every invocation. However, at ~160 inline lines with no one-level-deep reference split (e.g., for the detailed part-type catalog), it sits at 'good structure, minor organization gaps' rather than the ideal overview-plus-references shape.

4 / 5

Total

18

/

20

Passed

Description

92%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 description that clearly states what the skill does and when to use it, with concrete trigger scenarios and explicit differentiation from related built-in tools. The only weakness is modest synonym coverage in its trigger terms.

DimensionReasoningScore

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', 'locating specific past commands/decisions' — plus what it provides ('database schema, ready-to-use SQL query templates, and per-goal strategies'), giving comprehensive coverage of the skill's capabilities.

5 / 5

Completeness

Explicitly answers both: what ('Query the raw trajectory SQLite database... 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... or locating specific past commands/decisions that text search cannot surface', plus 'when the built-in memory and history tools are insufficient').

5 / 5

Trigger Term Quality

Good natural phrases users would say ('repeated errors', 'past commands/decisions', 'what was actually executed', 'text search cannot surface'), but common synonyms like 'transcript', 'conversation log', or 'chat history' are absent, so coverage stops short of the comprehensive-synonyms anchor.

4 / 5

Distinctiveness Conflict Risk

Clear niche — direct SQL analysis of the raw trajectory database — explicitly differentiated from the 'memory' and 'history' tools and from text search, so it is unlikely to trigger for the wrong skill.

5 / 5

Total

19

/

20

Passed

Validation

100%

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/cli/src/skill/builtin/.bundle/memory-search/SKILL.md

Reviewed

Table of Contents

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.