hierarchical-agent-memory

Scoped CLAUDE.md memory system that reduces context token spend. Creates directory-level context files, tracks savings via dashboard, and routes agents to the right sub-context.

Quality

46%

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 ./skills/hierarchical-agent-memory/SKILL.md

Quality

Content

38%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

This skill reads more like a product README or marketing document than an actionable skill for Claude. It describes a system conceptually — what it does, what commands exist, what the output looks like — but provides no implementation: no scripts, no templates, no executable code for generating CLAUDE.md files, launching the dashboard, or performing audits. The examples show idealized mock outputs rather than concrete steps Claude can follow.

Suggestions

Add executable implementation: include actual scripts or code for generating CLAUDE.md files, the dashboard server, and audit checks — or reference bundle files containing them.

Replace mock agent output examples with concrete file templates showing exactly what a generated root CLAUDE.md and subdirectory CLAUDE.md should contain.

Add validation steps to the workflow: after 'go ham' setup, verify generated files exist and meet token budget constraints; after 'ham route', verify routing entries match actual directories.

Split implementation details (dashboard code, audit logic, template generation) into bundle files and reference them from the main skill with clear navigation links.

Dimension	Reasoning	Score
Conciseness	The skill includes some unnecessary sections like 'When to Use This Skill' with 5 bullet points that over-explain obvious use cases, and the 'How It Works' section explains concepts (like context routing) that Claude could infer. The examples section with mock agent output adds bulk. However, the commands table and best practices are reasonably tight.	3 / 5
Actionability	The skill describes what happens conceptually but provides no executable code, scripts, or concrete implementation details. The 'commands' are trigger phrases but there's no actual code behind them — no scripts to run, no file templates to generate, no implementation of the dashboard. The examples show idealized agent output rather than executable steps.	2 / 5
Workflow Clarity	Steps 1-3 provide a rough sequence (setup → routing → dashboard), and the commands table lists available actions. However, there are no validation checkpoints — no verification that generated CLAUDE.md files are correct, no feedback loop for when setup fails or routing is misconfigured. The 'ham audit' command exists but isn't integrated into the workflow as a validation step.	3 / 5
Progressive Disclosure	No bundle files are provided despite the skill referencing complex functionality (dashboard, audit, routing, baseline capture) that would clearly benefit from separate implementation files. Everything is inlined in a single file with no references to supporting documentation. The 'Related Skills' section references other skills but the core skill lacks any structural decomposition.	2 / 5
	Total	10 / 20 Passed

Description

53%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 identifies a clear niche (scoped CLAUDE.md memory management) and lists several concrete capabilities, but lacks an explicit 'Use when...' clause that would help Claude know when to select this skill. The trigger terms are somewhat technical and may not match how users naturally describe their needs.

Suggestions

Add an explicit 'Use when...' clause with natural trigger phrases, e.g., 'Use when the user wants to organize CLAUDE.md files across directories, reduce context window token usage, or manage project-level memory.'

Include more natural synonyms and variations users might say, such as 'memory management', 'reduce context size', 'split context files', or 'project memory organization'.

Expand the capability list slightly to clarify what 'routes agents to the right sub-context' means in practical terms, e.g., 'automatically loads relevant context based on working directory'.

Dimension	Reasoning	Score
Specificity	Names the domain (CLAUDE.md memory system) and lists a few concrete actions (creates directory-level context files, tracks savings via dashboard, routes agents to sub-context), but the actions are somewhat abstract and not fully comprehensive.	3 / 5
Completeness	Has a clear 'what' (creates directory-level context files, tracks savings, routes agents) but no explicit 'when' clause. There is no 'Use when...' guidance, which per the rubric caps completeness at 3.	3 / 5
Trigger Term Quality	Includes some relevant terms like 'CLAUDE.md', 'context token', 'dashboard', and 'directory-level context files', but misses natural user phrases like 'memory management', 'reduce tokens', 'context splitting', or '.claude.md'. Users might not naturally say 'scoped memory system' or 'sub-context'.	3 / 5
Distinctiveness Conflict Risk	The description is fairly distinctive with its focus on CLAUDE.md files, directory-level scoping, and token savings tracking. There's minor overlap risk with general memory or context management skills, but the specific mention of CLAUDE.md and dashboard tracking makes it mostly distinct.	4 / 5
	Total	13 / 20 Passed

Validation

90%

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 — 10 / 11 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	10 / 11 Passed

Repository: administrakt0r/AI-Agents-Safe-Coding-Skills
Path: skills/hierarchical-agent-memory/SKILL.md
Commit: 57c135b

Reviewed: 4 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.