AgentIndex icon
AgentIndex
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompareGuides
Submit Tool
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompare
Home/
Memory & Context/
geml
geml logo

geml

Active·★ 29·NOASSERTION·Updated 2026-10-22

GEML is an Agent-Native markup format and protocol designed for collaborative document creation and editing between people and AI agents. It addresses critical issues like context bloat and data fragmentation by providing precise, addressable, and verifiable block-level operations.

#Markup Language#AI Agent Native#Document-as-a-Base#Context Management#AST Operations#LLM Tooling#Version Control
© 2026 AgentIndex.app|Built by a 10-year iOS Developer.
Best by use caseAboutContactAdvertisePrivacy PolicyTerms of UseQYSGitHubBuy me a coffee ☕

Browse by Category

MCP ServersCode AssistantWorkflow AutomationRAG / Knowledge BaseMulti-AgentBrowser AutomationLLM InfraDev ToolingObservability

Not affiliated with Anthropic, OpenAI or Microsoft.

Guides
#Data Serialization
$ Install
$ npm i -g @geml/geml
↗ Visit site★ GitHub
01

Features

01AST-level structured operations: Enables precise block-level create/read/update/delete, replacing whole-file rewrites with partial, idempotent updates.
02Low-token reads and writes: Significantly reduces prompt-token cost by allowing agents to operate on specific, addressable blocks, improving model throughput and lowering inference cost.
03Single source of truth with modular references: Supports native cross-document and cross-fragment references, ensuring changes at the source propagate everywhere and eliminate version skew.
04Robust two-way reads and writes: Utilizes a simple, uniform block structure well-suited for LLM generation, complemented by a strict validator providing precise error feedback.
02

Compatibility

Node.js
Runtime
Verified via docs
Claude Code
LLM Platform
Verified via docs
Python
Language
Verified via docs
VS Code
IDE
Verified via docs
Chrome
Browser
Verified via docs
Obsidian
Knowledge Base
Verified via docs
03

Quick start

1
$ npm i -g @geml/geml
04

Use cases

↳Enabling seamless, high-frequency collaboration between human writers and AI agents on shared documents.
↳Optimizing LLM context window usage and reducing token costs by allowing agents to operate on specific document blocks rather than entire files.
↳Maintaining document integrity, preventing semantic drift, and enabling traceability and atomic rollbacks in multi-round agent-driven document generation and editing.
↳Generating and dynamically visualizing code call graphs from entire codebases for debugging, impact analysis, and understanding code structure.
05

Alternatives

⚡Top VS:
VSclarilayer★77→VSfava-trails★26→VSratel★497→

Comments

Log in to leave a comment

No comments yet. Be the first!

On this page
01Features02Compatibility03Quick start04Use cases05Alternatives
VS
context-attachment-manager
★356
→
clarilayer logo
clarilayer★ 77
Stop re-explaining your data to your AI every session. The individual-analyst context layer, delivered over MCP (Claude Code / Cursor / Codex).
VSCompare→
fava-trails logo
fava-trails★ 26
🫛👣 FAVA Trails — Git-native, curated memory for AI agents via MCP. Draft isolation, promotion gate, thought lifecycle hooks, memory curation protocols, supersession chains.
VSCompare→
ratel logo
ratel★ 497
Context engineering for AI agents. ~80% fewer tokens. Fix tool overload. Skills and memory with in-process BM25 retrieval. No vector DB. No embeddings.
VSCompare→
context-attachment-manager logo
context-attachment-manager★ 356
Codex and Claude Code plugin (works in Claude Desktop): choose which past images go to the model with your next message. Unchecked images become placeholders, the conversation goes on, and requests stay small.
VSCompare→
ditto logo
ditto★ 298
Mine your Claude Code and Codex logs into a local you.md agent profile.
VSCompare→
Vault-Agent-Memory logo
Vault-Agent-Memory★ 50
Local-first memory governance for AI agents: shared, reviewable, auditable memory via SQLite and MCP.
VSCompare→
kaeru logo
kaeru★ 45
Cognitive memory for LLM agents — local-first, and shareable across a team through the cloud.
VSCompare→
rlm-claude logo
rlm-claude★ 39
Recursive Language Models for Claude Code - Infinite memory solution inspired by MIT CSAIL paper
VSCompare→
See all alternatives →
geml-spec /
geml
↗

One format, two readers. People and AI agents now co-write the same document. Legible for people; addressable, verifiable, and versioned for machines. GEML is plain text — organized by one typed block for everything, remembered by a .gemlhistory sidecar.

29
Last commit1d ago
StatusActive
CategoryMemory & Context
Links
Documentation↗Discussion↗Issues↗Releases↗
Cloud Infrastructure
Ad

Deploy on DigitalOcean

Host your MCP servers, Docker containers, and AI agents with $200 free credit over 60 days.

Claim $200 Free Credit