CtrlK
BlogDocsLog inGet started
Tessl Logo

engram-memory

ALWAYS ACTIVE — Persistent memory protocol. You MUST save decisions, conventions, bugs, and discoveries to engram proactively. Do NOT wait for the user to ask.

51

Quality

55%

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

Fix and improve this skill with Tessl

tessl review fix ./plugin/claude-code/skills/memory/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

67%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-structured and highly actionable with concrete tool names, a precise mem_save format spec, and clearly sequenced memory-recovery flows. Its main weaknesses are a duplicated fallback passage and the absence of a save-verification feedback loop in the proactive-save workflow.

Suggestions

De-duplicate the engram-setup fallback (currently stated both in AVAILABLE TOOLS and AFTER COMPACTION) into a single referenced location to tighten conciseness.

Add an explicit save→verify feedback step (e.g. confirm mem_save succeeded before proceeding, or retry on failure) to the proactive-save workflow to strengthen workflow clarity.

Consider extracting the session-close summary template and deferred-tool list into a reference file so the SKILL.md overview stays leaner.

DimensionReasoningScore

Conciseness

The body is largely efficient with sectioned trigger lists and a tight mem_save format spec, but it repeats the engram-setup fallback verbatim (in AVAILABLE TOOLS and again in AFTER COMPACTION) and restates the trigger categories in the self-check quote block, fitting the 'mostly efficient but includes some unnecessary explanation or could be tightened' anchor.

3 / 5

Actionability

It gives concrete, executable guidance: exact tool names, a full mem_save field spec (title/type/scope/topic_key/content with What/Why/Where/Learned), a session-summary template, and the exact fallback command 'engram setup claude-code', with only minor gaps (no filled example values and bracketed template placeholders), matching the 'mostly executable guidance with minor gaps' anchor.

4 / 5

Workflow Clarity

Multi-step flows are clearly sequenced with checkpoints — 'When to search memory' is a 1-2-3 escalation and 'After compaction' is an explicit 1-2-3 with 'IMMEDIATELY' and 'Only THEN continue' gates — but the proactive-save section is a trigger list rather than a sequenced flow and there is no save→verify feedback loop, fitting the 'clear sequence with most checkpoints; minor validation gaps' anchor.

4 / 5

Progressive Disclosure

The single ~125-line file is organized into well-labeled sections (AVAILABLE TOOLS, PROACTIVE SAVE TRIGGERS, WHEN TO SEARCH MEMORY, DELIVERY GUARANTEE, SESSION CLOSE PROTOCOL, AFTER COMPACTION) with no broken or deeply nested references, but it exceeds the under-50-line simple-skill threshold and inlines the session template and tool lists that could be split out, matching the 'good structure; most content appropriately placed; minor organization gaps' anchor.

4 / 5

Total

15

/

20

Passed

Description

42%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 clearly states what the skill does (proactive persistent-memory saving to engram) but is weak on natural user trigger language and leans on an always-on policy instead of explicit invocation guidance. Adding a 'Use when...' clause with user-natural phrases would lift completeness and trigger quality.

Suggestions

Add an explicit 'Use when the user says remember/recall/what did we decide, or after any decision, bug fix, or convention is established' clause to satisfy the completeness 'when' requirement and lift the cap at 3.

Include natural user-facing trigger phrases (e.g. 'remember this', 'save this', 'what did we do last time') so trigger-term quality reflects what users actually say rather than internal save categories.

Tighten the always-on framing or scope it to reduce overlap risk with generic memory/coding skills, making the niche more clearly distinct.

DimensionReasoningScore

Specificity

Names the domain (persistent memory) and several concrete save objects ('decisions, conventions, bugs, and discoveries'), but the action surface is essentially a single verb (save) without broader capability breadth, fitting the '1-2 concrete actions' anchor rather than a comprehensive multi-action list.

3 / 5

Completeness

It has a clear 'what' (save decisions/conventions/bugs/discoveries to engram), but the 'when' is only the degenerate always-on policy 'ALWAYS ACTIVE... Do NOT wait for the user to ask' rather than an explicit user trigger clause, so per the missing-'Use when...'-caps-at-3 guideline it lands at the 'clear what but when missing or weakly implied' anchor.

3 / 5

Trigger Term Quality

The terms present ('decisions, conventions, bugs, discoveries') are internal save categories, not natural phrases a user would say to invoke the skill, and there are no user-facing trigger phrases like 'remember this' or 'what did we decide' — matching the 'one or two generic keywords; missing the natural phrases users say' anchor.

2 / 5

Distinctiveness Conflict Risk

Naming the specific system ('engram') gives it a niche, but the always-on scope plus broad generic categories (decisions, bugs, conventions) creates overlap risk with general coding/memory skills, fitting the 'somewhat specific but could still overlap' anchor rather than the clearly-distinct anchor.

3 / 5

Total

11

/

20

Passed

Validation

100%

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

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
Gentleman-Programming/engram
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.