CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

search

Search memories from earlier Cursor sessions in this repository. Use it when earlier work may already explain the code, error, decision, or command you need, so you can avoid repeating file reads, searches, or experiments.

Quality

87%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

The canonical home for this skill is search in mem0ai/mem0

SKILL.md
Quality
Evals
Security

Quality

Content

92%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 exemplary short skill body: lean, concrete, and unambiguous, with tool-specific semantics (scopes, defaults, run_id filtering) explained in minimal space. The only gap is the absence of a worked example of the tool call with arguments.

DimensionReasoningScore

Conciseness

The ~18-line body is lean and every sentence carries non-obvious information (tool name, argument routing, scope semantics, run_id guard). It explains only tool-specific configuration Claude cannot know — e.g. "this repository's shared memory, which everyone who works in it contributes to" — not general concepts, matching the level-5 anchor 'lean and efficient; assumes Claude's competence; every token earns its place'.

5 / 5

Actionability

Concrete instruction is given: "Call `search_memories` with the user's question", with explicit per-argument handling ("Treat `--top-k`, `--category`, `--scope`, and `--run-id` as tool arguments") and default behavior for each omitted argument. It falls short of level 5's copy-paste-ready examples because no example invocation (e.g. how a question plus arguments look in the actual call) is shown, but it clearly exceeds level 3's incomplete/pseudocode level.

4 / 5

Workflow Clarity

This is a simple single-task skill and the single action is unambiguous: call the tool with the question, route flags as arguments, return the tool's result directly. Per the simple-skill exception, a single unambiguous action with no destructive or batch operations scores 5; there is nothing to sequence or validate.

5 / 5

Progressive Disclosure

The skill is under 50 lines, self-contained, and needs no external references — no references/, scripts/, or assets/ bundle exists and none is needed. Content is appropriately kept inline at this size, matching the guideline that short skills with well-organized, complete inline content score 5.

5 / 5

Total

19

/

20

Passed

Description

78%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: concrete what-and-when structure with explicit triggers and a well-scoped niche. Its main limits are that it names only one action and could include a few more natural trigger synonyms.

Suggestions

Name the kinds of content searched (e.g. code, errors, decisions, commands) in the 'what' half as concrete capabilities to lift specificity.

Add one or two natural trigger synonyms users would say, such as 'remember', 'recall', or 'Cursor history', to broaden trigger coverage.

DimensionReasoningScore

Specificity

The description names the domain ("memories from earlier Cursor sessions in this repository") and one concrete action ("Search"), but no other actions are listed. This matches the anchor 'Names domain and 1-2 concrete actions, but not comprehensive' — it does not reach level 4's 'several specific actions' since searching is the sole action named, and exceeds level 2 because the domain is precisely scoped rather than generic.

3 / 5

Completeness

It explicitly answers both questions: what — "Search memories from earlier Cursor sessions in this repository"; when — "Use it when earlier work may already explain the code, error, decision, or command you need". Both halves are concrete with explicit trigger phrases, matching the level-5 anchor; it is clearly above level 4, where the 'when' clause would be less specific.

5 / 5

Trigger Term Quality

Natural trigger words are present: "code, error, decision, or command you need" plus "file reads, searches, or experiments" — phrases a user would plausibly say. Coverage is good but misses common variations and synonyms (e.g. "remember", "previously done", "Cursor history"), so it fits level 4 ('good keyword coverage; a few natural terms missing') rather than level 5's comprehensive synonym/extension coverage.

4 / 5

Distinctiveness Conflict Risk

The niche is clear and well-scoped (session memories specific to Cursor and this repository), making confusion with unrelated skills unlikely. Minor overlap risk remains with general codebase-search or memory-recall skills because triggers like "code" and "searches" are broad, fitting level 4 ('mostly distinct; minor overlap risk') rather than level 5's minimal conflict risk.

4 / 5

Total

16

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository

mem0ai/mem0

Path

integrations/cursor-plugin/skills/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.