# Letta (slim)

> Letta, from the MemGPT team, is an open-source agent harness (formerly Letta Code) for stateful agents that keep memory blocks, skills and conversations across sessions, used from a CLI, chat.letta.com, messaging channels or the Agent SDK, with Letta Cloud.

- Full: https://www.anchorterminal.com/tools/letta.md (~7,450 tokens) · this version ~1,280 tokens · JSON https://www.anchorterminal.com/tools/letta.json · canonical https://www.anchorterminal.com/tools/letta
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · MCP: https://www.anchorterminal.com/mcp · Updated: 2026-11-07 · pricing https://www.anchorterminal.com/tools/letta#pricing · reviews https://www.anchorterminal.com/tools/letta#reviews
- Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data)

**C · 59.7/100 · rank #687 of 1510 · #18 in Agent harnesses · not agent-ready · confidence medium**


Assessment: Letta's open-source harness gives agents git-tracked memory blocks, skills, subagents and headless JSON runs, with a $0 plan and public per-agent API prices. It starts in the unrestricted permission mode, defaults to Letta Cloud, sends usage telemetry unless LETTA_CODE_TELEM=0 is set, and the V1 API server the comparisons describe is retired.

## Facts

- Kind: Agent harness · vendor: Letta · category: Agent harnesses · legal entity: not named · provenance 54/100
- Local only (stdio): npm `@letta-ai/letta-code`, pypi `letta`
- Auth: OAuth or key · pricing: Freemium · x402: no · licence: Apache-2.0
- Benchmark probes: not run yet, so Performance is pending
- Surface listed: The open-source Letta harness, `@letta-ai/letta-code` 0.34.42, with its CLI, headless mode and Agent SDK, read from clones of github.com/letta-ai/letta-code and letta-ai/letta, npm, PyPI, docs.letta.com's llms.txt and pricing page and letta.com's privacy policy on 6 November 2026. Nothing was installed or run
- Category: Agent harnesses, because Letta is now a finished program that runs the agent loop; memory is its distinguishing feature, so it also appears in agent memory
- Retired: The Letta V1 Python API server (letta-ai/letta, archive branch), its Docker images, the Electron desktop app and AgentFile import and export
- Interfaces: CLI and terminal UI, chat.letta.com in the browser, Slack, Telegram, Discord and custom channels, the TypeScript Agent SDK, the Letta App Server over WebSocket, and ACP
- Legal entity: Not named in the privacy policy, which says only Letta. The company announced itself as Letta, of San Francisco, in its September 2024 release
- Prices: Pro plan $20 per month (plan); API plan active agent $0.10 per month (plan)

- Scores: Reliability 58, Performance pending, Schema & documentation 70, Agent ergonomics 67, Security & auth 35, Payments & pricing 60, Task success pending, Maintenance & community 85, Transparency & trust 55 · total over the 7 assessed categories
- Why: Reliability, Local-software lines, read for the harness. · Schema & documentation, Harness reading. · Agent ergonomics, Harness reading. · Security & auth, Harness reading of the framework checklist, as for the other harnesses. · Payments & pricing, No x402, MPP or L402 (0 of 40). · Maintenance & community, v0.34.42 on 6 November 2026 (30). · Transparency & trust, Apache-2.0 (30).
- Sources: 8, open questions: 6, both in the full twin
- Capabilities: agent.harness, agent.mcp-client, agent.multi-agent, memory.store, memory.search
- JSON: https://www.anchorterminal.com/api/v1/tools/letta.json
- Verify (for the vendor): the badge `https://www.anchorterminal.com/badges/letta.svg` or a link to https://www.anchorterminal.com/tools/letta from a page on letta.com or one of its subdomains, or the README at the root of github.com/letta-ai/letta-code, then `POST https://www.anchorterminal.com/api/v1/verify` `{"slug", "url"}` or `verify_listing` at /mcp; re-checked weekly, no effect on the grade. Snippets in the full twin.
- Support: Discord https://discord.gg/letta; support@letta.com; https://docs.letta.com; https://github.com/letta-ai/letta-code/issues; more in the full twin

## Before you call it

1. Start headless runs with an explicit permission mode such as `standard` or `strict`; the default auto-approves tools
2. Pass `--backend local` for work that must stay on the machine, since Letta Cloud is the default
3. Set LETTA_CODE_TELEM=0 in unattended runs if usage events must not be sent
4. Use `letta -p --output-format json` for scripted runs and parse the JSON, not the terminal output
5. Don't plan on the retired V1 REST server or `.af` AgentFile import; both are gone

## Connect

```bash
npm install -g @letta-ai/letta-code
```

## Similar tools

| Tool | Grade | Score | Shared capabilities | Slim |
| --- | --- | --- | --- | --- |
| goose | BB | 71.3 | agent.harness, agent.mcp-client, agent.multi-agent | https://www.anchorterminal.com/tools/goose.min.md |
| AXLR | BB | 70.6 | agent.harness, agent.mcp-client, agent.multi-agent | https://www.anchorterminal.com/tools/axlr.min.md |
| OpenHands | BB | 70.3 | agent.harness, agent.mcp-client, agent.multi-agent | https://www.anchorterminal.com/tools/openhands.min.md |
| OpenCode | B | 68.8 | agent.harness, agent.mcp-client, agent.multi-agent | https://www.anchorterminal.com/tools/opencode.min.md |
| Qwen Code | B | 68.7 | agent.harness, agent.mcp-client, agent.multi-agent | https://www.anchorterminal.com/tools/qwen-code.min.md |


## Panel reviews (0, desk reviews from public material, no calls made)