memory

AI DevKit · Use the memory CLI as a durable knowledge layer. Search before non-trivial work, store verified reusable knowledge, update stale entries, and avoid saving transcripts, secrets, or one-off task progress.

Quality

84%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

88%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 is a well-engineered, actionable skill: fully executable commands, a sequenced workflow with an explicit validation gate and feedback loop, and clean sectioning. The only real weaknesses are minor conciseness redundancy between the gate and the store/don't-store lists and a length that slightly exceeds the simple-skill threshold for progressive disclosure.

Suggestions

Deduplicate the Quality Gate checklist and the Store/Do-not-store lists so each rule appears once, trimming tokens without losing guidance.

Consider whether the Commands reference (search/store/update flag tables) belongs in a separate REFERENCE.md to shorten SKILL.md below the simple-skill threshold, which would lift progressive_disclosure toward 5.

Keep the explicit Quality Gate and conflict-handling feedback loop — these are the strongest part of the workflow.

Dimension	Reasoning	Score
Conciseness	The body is lean — executable commands in code blocks, tight bullet checklists, no explanation of what memory is or how CLIs work — but there is mild redundancy between the Quality Gate checklist and the Store/Do-not-store lists that both restate "verified reusable knowledge" and "no secrets", fitting the score-4 anchor of efficient content with minor trim opportunities rather than 5's every-token-earns-its-place.	4 / 5
Actionability	Every command is fully executable and copy-paste ready — `npx ai-devkit@latest memory search/store/update` with all flags, the `--table` pattern for retrieving IDs, and concrete troubleshooting steps — covering the common cases exactly as the score-5 anchor requires.	5 / 5
Workflow Clarity	The workflow is clearly sequenced (search → use as context → search before storing → store/update after the gate) with an explicit Quality Gate validation checklist before persistence and a feedback loop ("If memory conflicts with verified evidence, use the evidence and update the stale memory"), matching the score-5 anchor's clear sequence with explicit validation and error-recovery loops; the destructive-operation cap does not apply because a strong validation gate is present.	5 / 5
Progressive Disclosure	The skill is a single self-contained file with no bundle files present and is organized into clear, logically ordered sections (Workflow, Quality Gate, Commands, Scoping, Troubleshooting), fitting the score-4 anchor of good structure with content appropriately placed; it does not reach 5 because there are no one-level-deep references to signal and the ~120-line length exceeds the "under 50 lines" simple-skill exception that would allow a 5 from structure alone.	4 / 5
	Total	18 / 20 Passed

Description

80%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 strong: third-person imperative voice, comprehensive concrete actions, and a clear product-scoped niche. Its main weakness is a generic "when" trigger and limited synonym coverage, which cap completeness and trigger-term quality at 4 rather than 5.

Suggestions

Replace the generic "before non-trivial work" trigger with concrete trigger phrases, e.g. "Use when implementing, debugging, reviewing, planning, or documenting tasks."

Add natural synonyms a user might say, such as "remember", "recall", or "persist context", to broaden trigger-term coverage.

Keep the third-person voice and concrete action list as-is — these are already strong.

Dimension	Reasoning	Score
Specificity	The description lists multiple concrete actions — "Search before non-trivial work", "store verified reusable knowledge", "update stale entries", "avoid saving transcripts, secrets, or one-off task progress" — giving comprehensive coverage of the memory CLI's capabilities, matching the score-5 anchor rather than the 4 anchor which allows minor gaps.	5 / 5
Completeness	It answers both "what" (use the memory CLI as a durable knowledge layer; search/store/update) and "when" ("before non-trivial work"), but the trigger is generic rather than concrete phrases like "when implementing, debugging, or reviewing", placing it at the score-4 anchor where the "when" could be more explicit rather than 5's concrete trigger phrases.	4 / 5
Trigger Term Quality	It includes solid natural keywords ("memory", "knowledge layer", "search", "store") but misses common synonyms a user might say such as "remember", "recall", or "context", fitting the score-4 anchor of good coverage with a few natural terms missing rather than comprehensive synonym coverage at 5.	4 / 5
Distinctiveness Conflict Risk	The product-specific prefix "AI DevKit ·" and the narrow memory/knowledge-layer purpose make it mostly distinct with only minor overlap risk against other knowledge-management skills, fitting the score-4 anchor rather than the 5 anchor that requires minimal conflict risk.	4 / 5
	Total	17 / 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: codeaholicguy/ai-devkit
Path: skills/memory/SKILL.md
Commit: d56e554

Reviewed: 3 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.