AgentIndex icon
AgentIndex
ToolsCategoriesTrendingNewCompare
Submit Tool
Home/
Memory & Context/
dory
dory logo

dory

Active·★ 11·MIT·Updated 2026-05-24
★ Trending★ Memory & Context★ LLM Infra

One memory layer for every AI agent. Local-first, markdown source of truth, and CLI/HTTP/MCP native. Your agent forgot who you are. Again. Dory fixes that.

Dory is a local-first memory daemon that gives every AI agent the same brain. It uses Markdown as source of truth and SQLite as a disposable sidecar, providing a narrow API for agents to read and write shared memory. This eliminates the need to re-explain context across different tools like Claude, Codex, opencode, etc.

#agents#ai-agents#claude-code#codex#docker#fastapi#knowledge-graph#llm
$ Install
$ git clone <dory-repo-url> && cd dory && uv sync --frozen && mkdir -p data/corpus && export DORY_CORPUS_ROOT="$PWD/data/corpus" && export DORY_INDEX_ROOT="$PWD/.dory/index" && export DORY_AUTH_TOKENS_PATH="$PWD/.dory/auth-tokens.json" && uv run dory init
↗ Visit site★ GitHub
01

Features

01Local-first memory daemon with Markdown as source of truth
02Narrow API: wake, search, get, memory-write, link
03Native MCP support and HTTP daemon
04Supports multiple AI agents (Claude, Codex, opencode, etc.)
05Deployable via Docker, CLI, or as a durable service
02

Compatibility

MCP
Native MCP
Verified via docs
HTTP
HTTP API
Verified via docs
Claude
Claude Support
Verified via docs
Codex
Codex Support
Verified via docs
OpenClaw
OpenClaw Package
Verified via docs
03

Quick start

1
$ git clone <dory-repo-url>
2
$ cd dory
3
$ uv sync --frozen
4
$ mkdir -p data/corpus
5
$ export DORY_CORPUS_ROOT="$PWD/data/corpus"
6
$ export DORY_INDEX_ROOT="$PWD/.dory/index"
7
$ export DORY_AUTH_TOKENS_PATH="$PWD/.dory/auth-tokens.json"
8
$ uv run dory init
04

Use cases

↳Persistent memory across different coding agents (Claude, Codex, etc.)
↳Session context recovery and cross-tool state management
↳Human-readable audit trail via Markdown and git
05

Alternatives

letta logo
letta★ 23.0k
Letta is the platform for building stateful agents: open AI with advanced memory that can learn and self-improve over time.
vs →
DesktopCommanderMCP logo
DesktopCommanderMCP★ 6.1k
This is MCP server for Claude that gives it terminal control, file system search and diff file editing capabilities
vs →
fastmcp logo
fastmcp★ 25.4k
🚀 The fast, Pythonic way to build MCP servers and clients.
vs →
Auto-claude-code-research-in-sleep logo
Auto-claude-code-research-in-sleep★ 11.0k
ARIS ⚔️ (Auto-Research-In-Sleep) — Claude Code skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation via Codex MCP
vs →
agents-best-practices logo
agents-best-practices★ 1.1k
Provider-neutral Agent Skill for Codex, Claude Code, and agentic harness design.
vs →
holaOS logo
holaOS★ 5.4k
The agent environment for long-horizon work, continuity, and self-evolution.
vs →
semble logo
semble★ 4.5k
Fast and Accurate Code Search for Agents
vs →
awesome-claude logo
awesome-claude★ 250
HeyClaude is a curated registry and distribution surface for Claude and AI-workflow assets: agents, MCP servers, skills, commands, hooks, rules, guides, tools, jobs, Raycast feeds, static data exports, and an npm MCP package.
vs →
See all alternatives →

Related searches

dory AlternativesBest Memory & Context Tools 2026Open Source Memory & Contextdory Tutorialdory Vs Competitorsagentsai-agentsclaude-code

Comments

Log in to leave a comment
  • Rebel Chen
    Rebel ChenMay 26, 2026

    Markdown as the source of truth for agent memory is the right call — readable, editable, not locked in a proprietary format.

  • S
    Sage LewisApr 19, 2026

    CLI/HTTP/MCP all supported — flexible enough to integrate with whatever agent architecture you're running.

  • M
    Marlowe KimApr 13, 2026

    MCP native support works cleanly. Plugged into an existing agent setup in under 10 minutes.

  • F
    Finley KimMar 7, 2026

    Finally a memory layer that doesn't require running a separate server or managing a cloud account.

  • Casey Kim
    Casey KimMar 4, 2026

    The local-first approach means memory actually persists between sessions without cloud dependencies.

On this page
01Features02Compatibility03Quick start04Use cases05Alternatives
Stats
GitHub Stars★ 11
Last commit6d ago
StatusActive
LicenseMIT
CategoryMemory & Context
Trend (30d)
+0.4↑ 0.7%
Links
Documentation↗Discussion↗Issues↗Releases↗

Deploy on DigitalOcean — Get $200 Free Credit

Ad
© 2026 AgentIndex.app|Built by a 10-year iOS Developer.
QYSGitHubBuy me a coffee ☕

Browse by Category

Code AssistantWorkflow AutomationRAG / Knowledge BaseMulti-AgentBrowser AutomationLLM InfraDev ToolingObservability

Not affiliated with Anthropic, OpenAI or Microsoft.