CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

claude-code-memory-and-rules

This skill should be used when the user asks to "set up memory", "configure CLAUDE.md", "add rules", "create project rules", "organize instructions", or mentions Claude Code memory, CLAUDE.md files, .claude/rules, rules, or project instructions. Also use when the user wants to persist preferences, add coding conventions, scope instructions to specific files, or understand when to use rules vs hooks or other memory mechanisms.

68

Quality

85%

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

82%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.

A well-organized, highly actionable reference skill: every recommendation resolves to a concrete path, template, or command, and navigation into the four reference files is clearly signaled. The main improvements are trimming known-concept explanation (glob semantics) and tightening the duplication between the inline rules/hierarchy sections and the corresponding reference files.

Suggestions

Trim or drop the Glob Patterns table — glob semantics are prior knowledge; keep only Claude-Code-specific behavior notes (e.g., how `paths` matching interacts with rule loading).

Condense the inline 'Modular Rules' section to the concept and one short example, leaving the directory-structure walkthrough and full YAML example to references/rules-guide.md, since they currently duplicate it nearly verbatim.

Add a short verification step after writing a rule or CLAUDE.md (e.g., 'Run /memory to confirm the file is loaded and rules are discovered') to close the workflow's feedback loop.

DimensionReasoningScore

Conciseness

The body is dense and earns its tokens — decision tree, hierarchy table, and templates with almost no prose padding (e.g., the "Choosing the Right Location" tree, the Rules vs Hooks table). Minor over-explanation remains: the Glob Patterns table explains pattern semantics ("**/*.ts | All TypeScript files") Claude already knows, and the Tips bullets restate the section's point.

4 / 5

Actionability

Fully executable guidance throughout: exact file paths for every scope, a copy-paste-ready CLAUDE.md template, a working `paths` frontmatter YAML block, `@path/to/file` import syntax with the max-depth constraint, and concrete commands (`/init`, `/memory`). The Rules vs Hooks table maps real example needs to specific locations. Common cases are covered with essentially no gaps.

5 / 5

Workflow Clarity

The core decision flow is unambiguous — the ASCII decision tree sequences need → scope → sharing → mechanism, and the rules-vs-hooks table handles the enforcement fork. This is a non-destructive configuration skill, so the validation-checkpoint cap does not apply, but there is also no verification step (e.g., 'confirm what's loaded via `/memory`' appears only as a bootstrapping aside), which keeps it at 'most checkpoints present' rather than fully explicit validation.

4 / 5

Progressive Disclosure

Good structure: a Quick Reference signals all four real, one-level-deep reference files (verified: no nested references in the bundle) with one-line descriptions, and the body works as an overview. It falls short of the top anchor because the inline 'Modular Rules' section (directory tree and path-scoped rules example) and the Memory Hierarchy table substantially duplicate references/rules-guide.md and references/memory-types.md nearly verbatim, so the split is not fully clean.

4 / 5

Total

17

/

20

Passed

Description

82%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.

A strong description with best-in-class trigger coverage and an explicit, multi-clause 'use when' structure. Its only structural weakness is that the capability statement ('what') is carried entirely by quoted user phrases rather than stated directly, leaving a small gap versus the top anchor.

DimensionReasoningScore

Specificity

Phrases like "configure CLAUDE.md", "add rules", "persist preferences", and "scope instructions to specific files" list several concrete, specific actions. It stops short of a 5 because the actions are framed as quoted user requests rather than an independent statement of what the skill does, and coverage of capabilities (e.g., editing or reviewing existing memory setups) is implied rather than listed.

4 / 5

Completeness

The 'when' is fully explicit ("should be used when the user asks to...", "Also use when the user wants to...") with concrete trigger phrases. The 'what' is present but only through the quoted trigger actions, not a direct capability statement like 'Configures Claude Code's memory system', so it fits the anchor where both exist but one could be more explicit.

4 / 5

Trigger Term Quality

Comprehensive natural-language coverage with synonyms and exact technical tokens: "set up memory", "configure CLAUDE.md", "add rules", "create project rules", "organize instructions", "Claude Code memory", "CLAUDE.md files", ".claude/rules", "persist preferences", "add coding conventions". These are the phrases a user would naturally say; no common variation is missing.

5 / 5

Distinctiveness Conflict Risk

Distinctive anchors ("Claude Code memory", "CLAUDE.md files", ".claude/rules") carve a clear niche with minimal conflict. The bare terms "rules", "add rules", and "organize instructions" are generic enough to risk overlap with unrelated rule-ish skills (linting rules, other Claude Code config skills), keeping it below the minimal-conflict anchor.

4 / 5

Total

17

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

allowed_tools_field

'allowed-tools' contains unusual tool name(s)

Warning

Total

15

/

16

Passed

Repository

dwmkerr/claude-toolkit

Path

plugins/toolkit/skills/claude-code-memory-and-rules/SKILL.md

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.