CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

memory-recall

Search and recall relevant memories from past sessions via memsearch. Use when the user's question could benefit from historical context, past decisions, debugging notes, previous conversations, or project knowledge -- especially questions like 'what did I decide about X', 'why did we do Y', or 'have I seen this before'. Also use when you see `[memsearch] Recall available if needed` capability hints injected via SessionStart or UserPromptSubmit. Typical flow: search for 3-5 chunks, expand the most relevant, optionally deep-drill into original transcripts via the anchor format. Skip when the question is purely about current code state (use Read/Grep), ephemeral (today's task only), or the user has explicitly asked to ignore memory.

Quality

94%

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

85%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 an executable, well-sequenced retrieval workflow with explicit fallbacks for the tool's known failure modes. Its only weaknesses are minor: a duplicated uvx fallback note, slightly padded fallback explanations, and one unresolved placeholder in the collection-derivation command.

Suggestions

State the 'uvx memsearch' fallback once (e.g., in a preliminary note) instead of repeating it in both step 1 and step 4.

Replace or resolve the '__DERIVE_COLLECTION_SCRIPT__' placeholder with the actual script path or a complete inline command so the collection-derivation step is copy-paste executable.

Tighten the step-3 fallback prose — drop justification like 'This avoids the Milvus lock file issue' and keep just the condition and the alternative command.

DimensionReasoningScore

Conciseness

The body is efficient with concrete commands and no explanations of concepts Claude already knows, but the 'If memsearch is not found, use uvx memsearch' note is repeated in steps 1 and 4, and fallback prose ('This avoids the Milvus lock file issue') could be trimmed — matching 'minor instances of over-explanation that could be trimmed' rather than the lean 5 anchor.

4 / 5

Actionability

Concrete, runnable commands throughout (search with flags, expand, transcript, MDIR shell snippets), but the collection-derivation command contains the unresolved '__DERIVE_COLLECTION_SCRIPT__' placeholder, which is not copy-paste executable as written — a minor gap consistent with the 4 anchor.

4 / 5

Workflow Clarity

Clear 1–5 sequence (search → evaluate → expand → deep drill → return) with an explicit filtering checkpoint in step 2 and genuine error-recovery feedback loops (lock/permission fallback to reading the source file; unrecognized transcript format fallback to reading the referenced file).

5 / 5

Progressive Disclosure

Single-file skill under 50 lines with no external references needed; sections (Project Collection, Steps, When unsure, Output Format) are well-organized and every piece of content is appropriately inline, satisfying the simple-skill exception for a top score.

5 / 5

Total

18

/

20

Passed

Description

100%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 exemplary: it names concrete capabilities, includes natural trigger phrases with synonyms, explicitly covers both what and when plus when-not-to-use, and stakes out a distinct niche with low conflict risk. Every clause carries information with no padding.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — 'Search and recall relevant memories', 'search for 3-5 chunks, expand the most relevant, optionally deep-drill into original transcripts' — with comprehensive coverage of the tool's capability surface, matching the top anchor rather than the 'minor gaps' anchor at 4.

5 / 5

Completeness

Explicitly answers both 'what' (search/expand/deep-drill via memsearch) and 'when' ('Use when the user's question could benefit from historical context...') with concrete trigger phrases, and adds explicit skip conditions — clearly exceeds the 4 anchor where 'when' could be more specific.

5 / 5

Trigger Term Quality

Natural user phrasings like 'what did I decide about X', 'why did we do Y', 'have I seen this before' plus synonyms (past decisions, debugging notes, previous conversations, project knowledge) give comprehensive coverage of terms a user would actually say.

5 / 5

Distinctiveness Conflict Risk

Clear niche (memsearch memory recall) with a unique trigger ('[memsearch] Recall available if needed' capability hints) and an explicit skip clause ('Skip when the question is purely about current code state') that minimizes conflict with code-reading skills.

5 / 5

Total

20

/

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

zilliztech/memsearch

Path

plugins/codex/skills/memory-recall/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.