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

1.06x

Quality

29%

Does it follow best practices?

Impact

78%

1.06x

Average score across 3 eval scenarios

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./cli-tool/components/skills/ai-research/agent-memory-systems/SKILL.md

Quality

Content

22%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 a sparse skeleton with no actionable content: empty anti-pattern headers, one-line pattern stubs, and a malformed Sharp Edges table whose Solution column contains stray markdown headers. It tells Claude nothing concrete about how to execute any memory-system task.

Suggestions

Flesh out each Pattern with concrete decision criteria or executable steps (e.g., vector store selection factors, chunk-size testing commands), and give each Anti-Pattern a one-line explanation of the failure mode and fix.

Rebuild the Sharp Edges table so each row has a real Issue, Severity, and actionable Solution instead of stray '## ...' header fragments.

Remove the role-play preamble and add at least one concrete worked example (e.g., a chunking + embedding + retrieval snippet with metadata filtering) so the skill is actionable rather than descriptive.

Dimension	Reasoning	Score
Conciseness	The body is short overall, but the opening role-play and truisms Claude already knows ('You are a cognitive architect...', 'Memory failures look like intelligence failures') are padded and could be trimmed.	3 / 5
Actionability	There is no executable guidance anywhere: Patterns are one-line abstractions ('Choosing the right memory type for different information'), Anti-Patterns have no body, and the Sharp Edges table is malformed with header fragments in the Solution column.	1 / 5
Workflow Clarity	No multi-step process is sequenced and there are no validation checkpoints; the Patterns are just labels and the Sharp Edges table is incoherent rather than a usable workflow.	1 / 5
Progressive Disclosure	Section headers exist (Capabilities, Patterns, Anti-Patterns, Sharp Edges) but organization is poor — the Sharp Edges table is broken, Anti-Pattern headers are empty, and no bundle files exist or are referenced.	3 / 5
	Total	8 / 20 Passed

Description

36%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 reads as a philosophical manifesto rather than a skill description: it is truncated mid-word, padded with abstract truisms, names no concrete actions, and provides no 'Use when' trigger guidance. It defines a niche but does not tell Claude when to invoke it or what it produces.

Suggestions

Replace the philosophical prose with concrete actions, e.g. 'Designs agent memory architectures: selects memory types (short-term, long-term), chooses vector stores, and defines chunking and retrieval strategies.'

Add an explicit 'Use when...' trigger clause naming natural user phrases, e.g. 'Use when the user asks about agent memory, RAG retrieval quality, vector store selection, or chunking strategies.'

Fix the truncated description (currently cut off mid-word at 'fragm') and remove padding such as 'Memory is the cornerstone of intelligent agents.'

Dimension	Reasoning	Score
Specificity	The description names the domain ('architecture of agent memory: short-term (context window), long-term (vector stores)') but uses abstract, philosophical language ('Memory is the cornerstone', 'Key insight: Memory isn't just storage - it's retrieval') with no concrete actions a user could map to a deliverable.	2 / 5
Completeness	It offers only a vague 'what' ('covers the architecture of agent memory') and entirely omits any 'when' / 'Use when' clause, matching the 'vague what and no when' anchor.	2 / 5
Trigger Term Quality	It includes several relevant domain keywords ('memory', 'retrieval', 'chunking', 'embedding', 'vector stores') but these are technical jargon rather than natural user phrasings, and there is no 'Use when...' trigger guidance.	3 / 5
Distinctiveness Conflict Risk	'Agent memory' is a distinct niche, but the retrieval/embedding/vector-store framing overlaps with RAG and embedding skills, and the absence of explicit triggers raises conflict risk.	3 / 5
	Total	10 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

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

	Total	15 / 16 Passed

Repository: davila7/claude-code-templates
Path: cli-tool/components/skills/ai-research/agent-memory-systems/SKILL.md
Commit: d4f034a

Reviewed: 5 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.