agent-memory-system

Persistent memory architecture for AI agents across sessions. Episodic memory (past events), procedural memory (learned skills), semantic memory (knowledge graph), short-term memory (active context). Use when implementing cross-session persistence, skill learning, context preservation, personalization, or building truly adaptive AI systems with long-term memory.

Quality

75%

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/agent-memory-system/SKILL.md

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 an over-long, concept-heavy reference that explains familiar memory taxonomy and relies on unimplemented pseudocode APIs and missing bundle files. It has structure but is verbose, only partially executable, and lacks validation checkpoints in its destructive batch workflows.

Suggestions

Trim the repeated memory-type explanations to a single concise section and drop concept padding Claude already knows, keeping only the novel architecture decisions and storage layouts.

Replace the fictional storeMemory/recallMemory APIs and CLI commands with real, executable code, or clearly mark them as illustrative interfaces and provide actual scripts in ./scripts/.

Add validation/verification checkpoints to the nightly consolidation and session-end hooks (e.g., verify archive integrity before deleting source episodes) and move large JSON schemas and appendices into ./references/ files linked one level deep.

Dimension	Reasoning	Score
Conciseness	The ~950-line body re-explains memory-type concepts Claude already knows and repeats the four types across Memory Types, Core Operations, Quick Reference, and integration sections, with padded emoji hook echoes and duplicated large JSON blocks.	2 / 5
Actionability	Concrete JSON schemas, file paths, and bash hooks exist, but the core storeMemory/recallMemory APIs and the recall-episode/get-skill/query-knowledge CLIs are pseudocode or fictional, and referenced scripts (examples/, .memory/scripts/*.js) are not provided as bundle files.	3 / 5
Workflow Clarity	Sequences (lifecycle, consolidation, session hooks) are present, but the nightly consolidation runs batch find-delete and mv-to-archive with no validation checkpoint, triggering the destructive/batch cap of 3.	3 / 5
Progressive Disclosure	Section headers give reasonable structure, but large JSON, multi-language code, and appendices that belong in separate files are inlined, and the referenced files (examples/, Appendix B, .memory/scripts/) do not exist as bundle files.	3 / 5
	Total	11 / 20 Passed

Description

83%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 is strong: third-person voice, explicit 'Use when' triggers, and a clear, structured statement of the four-memory-type architecture. It is comprehensive on completeness but slightly less concrete on actions and natural-language synonyms.

Dimension	Reasoning	Score
Specificity	Lists four memory types and five use-case capabilities (cross-session persistence, skill learning, personalization), giving several specific items; falls short of 5 because they are capability nouns rather than concrete verbs like 'store', 'recall', 'query'.	4 / 5
Completeness	Clearly states what the skill provides (memory architecture with four typed stores) and explicitly gives a 'Use when...' clause with concrete trigger phrases, satisfying both anchors.	5 / 5
Trigger Term Quality	Includes natural terms like 'persistent memory', 'cross-session persistence', 'skill learning', 'personalization', and 'long-term memory', but misses common variations users might say such as 'remember across chats' or 'memory store'.	4 / 5
Distinctiveness Conflict Risk	The persistent cross-session memory niche is distinct, but 'personalization' and 'skill learning' could overlap with related skills, so it is mostly distinct with minor overlap risk rather than minimal conflict.	4 / 5
	Total	17 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (958 lines); consider splitting into references/ and linking	Warning
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning

	Total	14 / 16 Passed

Repository: fernandezbaptiste/Skrillz
Path: skills/agent-memory-system/SKILL.md
Commit: 93ed392

Reviewed: 1 day 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.