recall

Search agentmemory for past observations, sessions, and learnings about a topic using hybrid BM25 plus vector plus graph search. Use when the user says "recall", "what did we do about", "did we ever", "have we seen", or needs context from past sessions.

Quality

91%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

100%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 lean skill body: concrete tool invocation with expected output, a sequenced workflow with edge-case and conflict-handling branches, WRONG/RIGHT anti-patterns, and a verification checklist. It adds only what Claude would not already know and keeps the sole external reference one level deep and clearly signaled.

Dimension	Reasoning	Score
Conciseness	Every section earns its tokens: a quick-start invocation with expected output, a 5-step workflow, terse anti-patterns, and a 4-item checklist, with no explanation of concepts Claude already knows ("Only surface what the tool returned. Never fabricate an observation, a session id, or an importance score"). Lean, assumes competence, nothing to trim.	5 / 5
Actionability	Fully concrete guidance: the exact call `memory_smart_search { "query": …, "limit": 10 }` with sample output, specific thresholds (`importance >= 7`), enumerated provenance channels (`user`, `agent`, `tool`, `import`, `shared`), and a verbatim RIGHT response for the empty-results case. Copy-adaptable and covers the common cases.	5 / 5
Workflow Clarity	Five numbered steps with explicit checkpoints: conflict resolution (prefer `user` over `agent` inference), the zero-results branch ("suggest 2-3 alternative search terms and stop. Do not guess"), and a closing checklist that verifies the outcome. Not a destructive or batch operation, so no validation cap applies.	5 / 5
Progressive Disclosure	No bundle files exist and the ~56-line body is a self-contained overview where nothing inlined belongs in a separate file; sections are well-organized and easy to navigate. The only external pointer (`../_shared/TROUBLESHOOTING.md`) is one level deep and clearly signaled, and 'See also' names sibling skills for navigation.	5 / 5
	Total	20 / 20 Passed

Description

83%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 with an explicit 'Use when' trigger clause, natural quoted trigger phrases, and a concrete statement of what is searched and how. The main gaps are missing common synonyms in the trigger list and mild overlap risk with sibling memory skills in the same family.

Suggestions

Add one or two natural synonyms to the trigger list (e.g., 'remember when', 'last time we', 'previously discussed') to strengthen trigger_term_quality toward comprehensive coverage.

Sharpen the trigger boundary against sibling skills, e.g., 'Use when the user asks about past work; for capturing new information, use remember instead' to reduce overlap risk with the write-side and session-recap skills.

Mention one downstream action beyond search (e.g., 'group results by session and lead with high-importance observations') to round out the capability list toward anchor-5 specificity.

Dimension	Reasoning	Score
Specificity	"Search agentmemory for past observations, sessions, and learnings about a topic using hybrid BM25 plus vector plus graph search" names the domain, a concrete action, the searchable objects, and the method — several specific elements with only minor gaps (single verb, post-search behavior undescribed). It exceeds anchor 3 ('not comprehensive') but falls short of anchor 5's multiple distinct concrete actions.	4 / 5
Completeness	Explicitly answers both what ("Search agentmemory for past observations, sessions, and learnings… using hybrid BM25 plus vector plus graph search") and when ("Use when the user says 'recall', 'what did we do about', 'did we ever', 'have we seen'…") with concrete trigger phrases, matching the anchor-5 pattern.	5 / 5
Trigger Term Quality	Triggers "recall", "what did we do about", "did we ever", "have we seen", "needs context from past sessions" are natural phrases a user would actually say, with good variety. Not 5 because common synonyms like "remember when", "last time", or "previously" are missing.	4 / 5
Distinctiveness Conflict Risk	A clear retrieval niche with distinct quoted triggers ("recall", "did we ever"), but "needs context from past sessions" is broad and the memory domain is shared with closely related sibling skills (remember, recap, session-history), creating minor overlap risk. More conflictable than anchor 5's 'minimal conflict risk'.	4 / 5
	Total	17 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: rohitg00/agentmemory
Path: plugin/skills/recall/SKILL.md
Commit: c314c7b

Reviewed: 4 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.