CtrlK
BlogDocsLog inGet started
Tessl Logo

agent-memory-mcp

A hybrid memory system that provides persistent, searchable knowledge management for AI agents (Architecture, Patterns, Decisions).

Install with Tessl CLI

npx tessl i github:sickn33/antigravity-awesome-skills --skill agent-memory-mcp
What are skills?

60

8.00x

Quality

41%

Does it follow best practices?

Impact

96%

8.00x

Average score across 3 eval scenarios

Optimize this skill with Tessl

npx tessl skill review --optimize ./skills/agent-memory-mcp/SKILL.md
SKILL.md
Review
Evals

Evaluation results

95%

63%

Automate Agent Memory System Onboarding

Memory system setup automation

Criteria
Without context
With context

Clone destination path

0%

100%

Compile step present

100%

100%

start-server project_id arg

0%

100%

start-server absolute path arg

0%

100%

npm install step

100%

100%

npm run start-server used

0%

100%

Correct working directory

44%

100%

Node.js v18 requirement

0%

0%

Without context: $0.4669 · 1m 26s · 22 turns · 19 in / 3,857 out tokens

With context: $0.3124 · 59s · 19 turns · 150 in / 3,000 out tokens

100%

100%

Knowledge Capture Utility for Microservices Migration

MCP memory tool API usage

Criteria
Without context
With context

memory_write key arg

0%

100%

memory_write type value

0%

100%

memory_write content arg

0%

100%

memory_write tags arg

0%

100%

memory_search query arg

0%

100%

memory_search type filter

0%

100%

memory_search tags filter

0%

100%

memory_read key arg

0%

100%

memory_stats invoked

0%

100%

Without context: $0.2366 · 1m · 10 turns · 10 in / 3,748 out tokens

With context: $0.3469 · 1m 34s · 16 turns · 317 in / 5,467 out tokens

93%

88%

Memory Knowledge Base Exploration for a New Team Member

Dashboard startup and memory exploration

Criteria
Without context
With context

start-dashboard command

0%

100%

start-dashboard absolute path

0%

100%

Dashboard port 3333

0%

100%

memory_stats invoked

0%

100%

memory_search used

0%

100%

memory_search type filter

0%

100%

memory_read used

0%

0%

Dashboard browser access documented

100%

100%

Without context: $0.3604 · 1m 41s · 13 turns · 62 in / 6,111 out tokens

With context: $0.6920 · 2m 47s · 22 turns · 70 in / 11,811 out tokens

Evaluated
Agent
Claude Code
Model
Claude Sonnet 4.6

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.