memory-system

Persistent cross-session memory management. Enables agents to remember user preferences, project conventions, and past decisions across different sessions using a structured MEMORY.md index and topic files.

Quality

61%

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

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/memory-system/SKILL.md

Quality

Content

71%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 well-organized, concise, and actionable with clear sequenced workflows and a safety checkpoint on the destructive prune operation; its main weakness is the lack of any reference files, leaving a long single-file skill where some content could be split out.

Suggestions

Move the detailed example index, topic-file template, and taxonomy tables into a references/ file (e.g., MEMORY_FORMAT.md) and link from SKILL.md to improve progressive disclosure.

Add an explicit verify step to Recall (e.g., confirm the referenced topic file exists before applying) to strengthen the workflow feedback loop.

Tighten the 'What NOT to Save' and Taxonomy tables by merging redundant columns to reclaim a few tokens.

Dimension	Reasoning	Score
Conciseness	The body is efficient: tables, compact examples, and a useful token-impact note, with only minor sections (What NOT to Save, Taxonomy) that could be trimmed; it largely assumes Claude's competence rather than over-explaining.	4 / 5
Actionability	Concrete formats and examples (index entries, topic-file frontmatter, the Save/Recall/Search/Prune step lists with explicit file paths) give mostly executable guidance, with only minor gaps (no literal shell commands, but file paths and formats are specific).	4 / 5
Workflow Clarity	Save/Recall/Search/Prune are clearly sequenced, and the destructive Prune op includes an explicit checkpoint ('Never auto-delete — always ask user first'); minor validation gaps (e.g., no index-integrity verify on recall) keep it just below 5.	4 / 5
Progressive Disclosure	No bundle/reference files exist and the entire skill is one ~183-line file; internal section structure is good, but large example and taxonomy blocks that could be split into references are inlined, with no one-level-deep external navigation.	3 / 5
	Total	15 / 20 Passed

Description

51%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 conveys a clear purpose and concrete artifacts but omits trigger guidance from the description field itself (delegating it to when_to_use), which limits trigger-term quality and completeness.

Suggestions

Fold a brief 'Use when the user says "remember this", "save for later", or when starting a session needing recalled context' clause into the description so it stands alone.

Add concrete verbs beyond 'remember' (e.g., 'save, recall, search, and prune cross-session memory') to raise specificity.

Include natural synonyms users say (e.g., 'notes', 'don't forget') directly in the description for better trigger-term coverage.

Dimension	Reasoning	Score
Specificity	Names concrete memory targets ('user preferences, project conventions, and past decisions') and concrete artifacts ('structured MEMORY.md index and topic files'), but the verbs are limited to 'remember'/'Enables', lacking the diverse distinct operations of a 5.	4 / 5
Completeness	A clear 'what' is present (persistent cross-session memory via MEMORY.md and topic files) but the description lacks any 'Use when...' clause; per guideline a missing trigger clause caps completeness at 3.	3 / 5
Trigger Term Quality	The description field contains only generic keywords ('memory', 'remember') and technical phrasing; the natural trigger phrases ('remember this', 'save this for later') live in the separate when_to_use field, not the description.	2 / 5
Distinctiveness Conflict Risk	The named artifacts (MEMORY.md index, topic files) and 'persistent cross-session memory' framing carve a mostly distinct niche with only minor overlap risk against generic note-taking skills.	4 / 5
	Total	13 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: vudovn/ag-kit
Path: .agents/skills/memory-system/SKILL.md
Commit: 211561c

Reviewed: 8 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.