agent-memory-systems

Memory is the cornerstone of intelligent agents. Without it, every interaction starts from zero. This skill covers the architecture of agent memory: short-term (context window), long-term (vector stores), and the cognitive architectures that organize them. Key insight: Memory isn't just storage - it's retrieval. A million stored facts mean nothing if you can't find the right one. Chunking, embedding, and retrieval strategies determine whether your agent remembers or forgets. The field is fragm

Quality

11%

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Passed

No known issues

Fix and improve this skill with Tessl

tessl review fix ./.agent/skills/agent-memory-systems/SKILL.md

Quality

Content

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

This skill is essentially an empty shell: a persona description, a tag list, and links to sub-skills that don't exist in the bundle. It contains no actionable instructions, no code examples, no workflows, and no concrete guidance. The verbose persona framing wastes tokens on concepts Claude already understands, while providing nothing Claude can actually execute.

Suggestions

Replace the persona narrative with a concise overview of memory system architecture (e.g., a decision table for when to use short-term vs. long-term memory, with concrete implementation patterns).

Add at least one executable code example showing a basic memory system setup (e.g., vector store initialization, chunking implementation, or retrieval query).

Define a clear workflow for designing an agent memory system: e.g., 1. Identify memory types needed → 2. Choose vector store → 3. Implement chunking → 4. Test retrieval quality → 5. Iterate on chunk size/overlap.

Either provide the referenced sub-skill files in the bundle or inline the essential content; currently the skill body has almost no standalone value.

Dimension	Reasoning	Score
Conciseness	The opening paragraphs are verbose persona-building that Claude doesn't need. Phrases like 'You are a cognitive architect who understands that memory makes agents intelligent' and the extended narrative about obsessing over chunking strategies are padding. The 'Capabilities' section is a list of tags with no actionable content. Very little of this earns its token cost.	1 / 3
Actionability	There is no concrete code, no executable commands, no specific examples, and no actual instructions. The content describes concepts abstractly and lists memory types but never tells Claude what to do. The body is entirely descriptive with zero actionable guidance.	1 / 3
Workflow Clarity	There are no steps, no sequences, no validation checkpoints, and no workflow of any kind. The skill provides no process for building, configuring, or debugging memory systems. It's a table of contents with no procedural content.	1 / 3
Progressive Disclosure	The skill does reference six sub-skill files with clear links, which is a reasonable progressive disclosure structure. However, no bundle files are provided to verify these references exist, the content sentence is cut off mid-thought, and the top-level overview provides almost no substantive content to orient the reader before diving into sub-skills.	2 / 3
	Total	5 / 12 Passed

Description

14%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

This description reads like an introductory essay or blog post about agent memory rather than a functional skill description. It lacks concrete actions, explicit trigger conditions, and appears to be truncated mid-sentence. The philosophical tone ('Memory is the cornerstone...') and editorial commentary ('Key insight...') add no value for skill selection and actively obscure what the skill actually does.

Suggestions

Replace the essay-style prose with concrete actions the skill performs, e.g., 'Designs agent memory architectures, configures vector stores, implements chunking and retrieval strategies for long-term agent memory.'

Add an explicit 'Use when...' clause with natural trigger terms, e.g., 'Use when the user asks about agent memory, RAG pipelines, vector databases, embedding strategies, or context window management.'

Remove philosophical commentary ('Memory is the cornerstone...', 'Key insight...') and ensure the description is complete (it currently appears truncated mid-word).

Dimension	Reasoning	Score
Specificity	The description uses abstract, conceptual language ('cornerstone of intelligent agents', 'cognitive architectures') rather than listing concrete actions. It mentions concepts like 'chunking, embedding, and retrieval strategies' but does not describe specific actions the skill performs (e.g., 'create vector stores', 'configure retrieval pipelines').	1 / 3
Completeness	The description partially addresses 'what' (covers agent memory architecture) but in vague, conceptual terms. There is no 'Use when...' clause or any explicit trigger guidance for when Claude should select this skill, and the description appears truncated.	1 / 3
Trigger Term Quality	It includes some relevant technical keywords like 'vector stores', 'context window', 'embedding', 'chunking', and 'retrieval' that a user might mention. However, it lacks common natural-language variations a user would say (e.g., 'RAG', 'knowledge base', 'semantic search') and reads more like an essay than a skill description.	2 / 3
Distinctiveness Conflict Risk	The description is very broad, covering memory, vector stores, embeddings, retrieval, and cognitive architectures. This could easily overlap with skills about RAG, embeddings, vector databases, or general AI agent design, making it highly conflict-prone.	1 / 3
	Total	5 / 12 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: Dokhacgiakhoa/antigravity-ide
Commit: c4a059b

Reviewed: 13 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.