CtrlK
BlogDocsLog inGet started
Tessl Logo

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.

74

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

86%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 actionable, well-structured retrieval workflow with copy-paste commands and sensible fallbacks; only minor conciseness trimming and an implicit (rather than explicit) validation checkpoint keep it from a perfect score.

Suggestions

Trim meta-instructions that restate Claude's own competence (e.g. "Be concise — only include information that is genuinely useful", "it is the source of truth for memory") to tighten the token budget.

Make the Evaluate step an explicit validation checkpoint (e.g. "If all top-k chunks are irrelevant, retry with a reformulated query before returning 'No relevant memories found'") to strengthen the workflow loop.

Consolidate the three repeated MDIR assignment snippets in the "When unsure" section into one defined variable used by all three commands to reduce redundancy.

DimensionReasoningScore

Conciseness

Mostly efficient and command-driven, but a few lines pad what Claude already knows ("it is the source of truth for memory", "Be concise — only include information that is genuinely useful") that could be trimmed without losing clarity.

4 / 5

Actionability

Fully executable, copy-paste-ready commands are given throughout — `memsearch search ... --top-k 5 --json-output`, `memsearch expand <chunk_hash>`, `memsearch transcript ... --turn <uuid> --context 3` — with a `uvx memsearch` fallback and concrete MDIR shell snippets.

5 / 5

Workflow Clarity

Steps 1-5 are clearly sequenced with an Evaluate filtering checkpoint and error-recovery fallbacks (tool missing → uvx, unrecognized format → read file directly), but it lacks an explicit validate/retry loop; read-only retrieval means no destructive-operation cap applies.

4 / 5

Progressive Disclosure

The body is a compact, single-purpose instruction set under ~50 lines with no bundle files to disclose, and is organized into clear sections (Project Collection, Steps, When unsure, Output Format), satisfying the simple-skill exception for a 5.

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.

A strong, third-person description that concretely states capabilities, provides rich natural-language triggers in the user's own voice, and explicitly covers both what and when with skip-when boundaries.

DimensionReasoningScore

Specificity

Names the tool (memsearch) and multiple concrete actions — "search for 3-5 chunks", "expand the most relevant", "deep-drill into original transcripts via the anchor format" — giving comprehensive coverage of what the skill does.

5 / 5

Completeness

Explicitly answers both what ("Search and recall relevant memories from past sessions via memsearch" + typical flow) and when ("Use when..." with concrete trigger phrases and a capability-hint injection path), and adds explicit skip-when guidance.

5 / 5

Trigger Term Quality

Comprehensive natural trigger coverage including direct user-voice quotes ("what did I decide about X", "why did we do Y", "have I seen this before") plus synonyms like "past decisions", "debugging notes", "previous conversations", and the literal capability-hint string.

5 / 5

Distinctiveness Conflict Risk

The memsearch/memory-recall niche is sharply defined by tool name, the `[memsearch] Recall available if needed` hint, and transcript-anchor mechanics, giving it distinct triggers with minimal overlap with other skills.

5 / 5

Total

20

/

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.

Validation15 / 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
zilliztech/memsearch
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.