CtrlK
BlogDocsLog inGet started
Tessl Logo

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.

69

Quality

86%

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

72%

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 and actionable with concrete templates, paths, and operation sequences. Its main weaknesses are minor front-matter redundancy and the absence of validation/verification feedback loops in the destructive Prune workflow.

Suggestions

Drop the redundant tagline/Overview restatement of the description and the 'Memory = what you KNOW…' closer to tighten token use.

Add an explicit verification step to Save (e.g., confirm the MEMORY.md pointer resolves to the created topic file) and a post-prune check to Prune so the destructive workflow has a feedback loop.

Consider whether the detailed index/topic-file format examples belong in a reference file once the skill grows, to keep SKILL.md as a lean overview.

DimensionReasoningScore

Conciseness

Mostly efficient via tables and templates, but the tagline, Overview, and Token Impact line restate the frontmatter description, and the philosophical 'Memory = what you KNOW…' closer is mild padding.

2 / 3

Actionability

Gives concrete, copy-paste-ready guidance — frontmatter and index-entry templates, exact glob/path targets like '.agents/memory/*.md', and scripted confirmation strings ('Saved to memory: [summary]').

3 / 3

Workflow Clarity

Save/Recall/Search/Prune are clearly numbered and Prune is gated by 'never auto-delete — ask user first', but there are no verify-after-write feedback loops and the destructive prune lacks an explicit post-action validation checkpoint.

2 / 3

Progressive Disclosure

No bundle files exist, yet the single SKILL.md is cleanly sectioned (Architecture, formats, taxonomy, Operations, Session Start Protocol) with self-contained overview-level content that needs no external splitting.

3 / 3

Total

10

/

12

Passed

Description

100%

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 third-person, specific, and pairs a clear capability statement with explicit, natural-language trigger guidance in when_to_use. It avoids fluff and over-claims while covering what, when, and how. No changes needed.

DimensionReasoningScore

Specificity

Lists multiple concrete capabilities — remembering 'user preferences, project conventions, and past decisions' via 'a structured MEMORY.md index and topic files' — naming both the domain and the concrete mechanism.

3 / 3

Completeness

Clearly answers both 'what' (persistent cross-session memory management via MEMORY.md + topic files) and 'when' (explicit when_to_use triggers), satisfying the explicit-trigger requirement.

3 / 3

Trigger Term Quality

The when_to_use field captures natural phrasing users actually say — 'remember this', 'save this for later', 'don't forget' — plus session-start and /remember invocation triggers.

3 / 3

Distinctiveness Conflict Risk

The MEMORY.md/topic-file mechanism and 'remember this'/'save this for later' triggers carve a clear niche unlikely to fire for plan or note-taking skills.

3 / 3

Total

12

/

12

Passed

Validation

87%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

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
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.