recollect

Resolve vague recall into recognized context through AI-guided contextual scan and user-validated recognition — one session, or the line of work, topic, or settled concept spread across several.

Quality

53%

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

Fix and improve this skill with Tessl

tessl review fix ./anamnesis/skills/recollect/SKILL.md

Quality

Content

52%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 protocol is structurally rigorous with strong workflow sequencing and well-organized external references, but it is markedly verbose through repetition of the same branch logic and leans heavily on abstract formalism over concrete, directly-actionable instruction.

Suggestions

Collapse the redundant restatements of branch logic across FLOW, PHASE TRANSITIONS, LOOP, and CONVERGENCE into a single canonical statement with cross-references, to cut substantial padding.

Move more of the dense FLOW/MORPHISM/TYPES formalism into a reference file, leaving SKILL.md as a leaner overview that points to it, which would also lift the progressive_disclosure and conciseness scores.

Lead the body with a concise actionable quick-start (the trigger, the reference-loading order, and the presentation shape) before the formal specification, so the most-used guidance is front-loaded.

Dimension	Reasoning	Score
Conciseness	The 338-line body restates the same branch logic across the FLOW, PHASE TRANSITIONS, LOOP, and CONVERGENCE blocks and carries heavy inline justification; while it avoids explaining concepts Claude already knows, the repetition and dense commentary are noticeable padding.	2 / 5
Actionability	The prose sections (reference loading, user-facing realization, rules) give concrete executable guidance, but the bulk of the body is abstract FLOW/MORPHISM/TYPES formalism that does not directly translate into action.	3 / 5
Workflow Clarity	Explicit Phases 0-3 are clearly sequenced with validation checkpoints (grounded/¬grounded, find_empty -> Ask -> recue -> Find, the Qx checkpoint, the attempts budget) and feedback loops for error recovery; the dense notation obscures readability but the structure is strong.	4 / 5
Progressive Disclosure	Eight references (claude, codex, entropy-track, salience-track, supra-session, fork-resume, failure-modes, round-composition) are well-signaled, one level deep, and all exist as real files, with detail appropriately split out; the only gap is that SKILL.md itself is a long formal wall rather than a lean overview.	4 / 5
	Total	13 / 20 Passed

Description

53%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 states a clear, distinctive purpose in third person, but its actions are somewhat abstract and it lacks an explicit 'Use when...' trigger clause. Trigger-term coverage is adequate but misses the most natural user phrasings.

Suggestions

Add an explicit 'Use when...' clause naming concrete trigger phrases (e.g., 'Use when the user says they vaguely remember a past discussion, can't recall a conversation, or refers to a line of work, topic, or settled concept').

Replace abstract verbs with more concrete operations so the actions read as specific capabilities rather than high-level intent.

Add common-sense synonyms and file/extension-style triggers users would naturally say, not just the specialized term 'recall'.

Dimension	Reasoning	Score
Specificity	Names the domain and concrete actions ('AI-guided contextual scan', 'user-validated recognition'), but they are stated at a fairly abstract level rather than as specific concrete operations like 'extract, fill, merge'.	3 / 5
Completeness	It clearly states the 'what' (resolve vague recall into recognized context) but has no 'Use when...' clause; the dash clause describes scope rather than an explicit trigger, capping completeness at 3 per the guidelines.	3 / 5
Trigger Term Quality	'vague recall' and 'recognized context' include the relevant keyword 'recall', but common natural phrasings users would actually say ('I vaguely remember', 'that conversation we had') are missing.	3 / 5
Distinctiveness Conflict Risk	The recall-resolution framing ('vague recall into recognized context', 'user-validated recognition') carves a distinct niche with only minor overlap risk against general memory-management skills. Voice is third person, so no specificity penalty applies.	4 / 5
	Total	13 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: jongwony/epistemic-protocols
Path: anamnesis/skills/recollect/SKILL.md
Commit: 3ffe94c

Reviewed: 1 day 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.