@letta-ai/letta-code 0.1.7 → 0.1.9
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +64 -5
- package/letta.js +3380 -1535
- package/package.json +8 -3
- package/vendor/ink-text-input/build/index.js +1 -7
package/README.md
CHANGED
|
@@ -2,12 +2,18 @@
|
|
|
2
2
|
|
|
3
3
|
A self-improving, stateful coding agent that can learn from experience and improve with use.
|
|
4
4
|
|
|
5
|
+
https://github.com/user-attachments/assets/5561a3ff-afd9-42a9-8601-55d245946394
|
|
6
|
+
|
|
7
|
+
---
|
|
8
|
+
|
|
5
9
|
## What is Letta Code?
|
|
6
10
|
|
|
7
11
|
Letta Code is a command-line harness around the stateful Letta [Agents API](https://docs.letta.com/api-reference/overview). You can use Letta Code to create and connect with any Letta agent (even non-coding agents!) - Letta Code simply gives your agents the ability to interact with your local dev environment, directly in your terminal.
|
|
8
12
|
|
|
13
|
+
Letta Code is model agnostic, and supports Sonnet 4.5, GPT-5, Gemini 2.5, GLM-4.6, and more.
|
|
14
|
+
|
|
9
15
|
> [!IMPORTANT]
|
|
10
|
-
> Letta Code is a **research preview** in active development, and may have bugs or unexpected issues. To learn more about the roadmap and chat with the dev team, visit our Discord
|
|
16
|
+
> Letta Code is a **research preview** in active development, and may have bugs or unexpected issues. To learn more about the roadmap and chat with the dev team, visit our [Discord](https:/discord.gg/letta). Contributions welcome, join the fun.
|
|
11
17
|
|
|
12
18
|
## Quickstart
|
|
13
19
|
|
|
@@ -28,12 +34,41 @@ Then run `letta` to start Letta Code (see various command-line options below):
|
|
|
28
34
|
letta
|
|
29
35
|
```
|
|
30
36
|
|
|
37
|
+
Any of the agents you create in Letta Code will be viewable (and fully interactable!) inside the [Agent Development Environment](https://app.letta.com).
|
|
38
|
+
|
|
39
|
+
## Persistence
|
|
40
|
+
|
|
41
|
+
All agents in Letta are **stateful**: they maintain context forever and can self-edit their own [memory blocks](https://www.letta.com/blog/memory-blocks). Agents can share memory blocks across projects—for example, multiple agents can share user coding preferences while maintaining project-specific memories independently.
|
|
42
|
+
|
|
43
|
+
### Memory Configuration
|
|
44
|
+
|
|
45
|
+
Letta Code uses a hierarchical memory system with both global and local blocks:
|
|
46
|
+
|
|
47
|
+
**Global** (`~/.letta/settings.json`)
|
|
48
|
+
- `persona` block - defines agent behavior
|
|
49
|
+
- `human` block - stores user coding preferences
|
|
50
|
+
|
|
51
|
+
**Local** (`./.letta/settings.json`)
|
|
52
|
+
- `project` block - stores project-specific context
|
|
53
|
+
|
|
54
|
+
### Starting Letta
|
|
55
|
+
|
|
56
|
+
```bash
|
|
57
|
+
letta # New agent (attaches to existing memory blocks or creates new)
|
|
58
|
+
letta --continue # Resume last agent session
|
|
59
|
+
letta --agent <id> # Resume specific agent session
|
|
60
|
+
```
|
|
61
|
+
|
|
62
|
+
When you start a new agent, it automatically connects to existing memory block IDs from your settings files. If none exist, it creates them.
|
|
63
|
+
|
|
64
|
+
Memory blocks are highly configurable — see our [docs](https://docs.letta.com/guides/agents/memory-blocks) for advanced configuration options. Join our [Discord](https://discord.gg/letta) to share feedback on persistence patterns for coding agents.
|
|
65
|
+
|
|
31
66
|
## Usage
|
|
32
67
|
|
|
33
68
|
### Interactive Mode
|
|
34
69
|
```bash
|
|
35
|
-
letta # Start new session
|
|
36
|
-
letta --continue # Resume last session
|
|
70
|
+
letta # Start new session (new agent with shared memory blocks)
|
|
71
|
+
letta --continue # Resume last session (last recently used agent)
|
|
37
72
|
letta --agent <id> # Open specific agent
|
|
38
73
|
```
|
|
39
74
|
|
|
@@ -57,6 +92,30 @@ letta -p "run tests" --tools "Bash,Read" # Only load specific tools
|
|
|
57
92
|
letta -p "analyze code" --tools "" # No tools (analysis only)
|
|
58
93
|
```
|
|
59
94
|
|
|
95
|
+
Use `--output-format json` to get additional information, including the agent ID ("session_id"):
|
|
96
|
+
```bash
|
|
97
|
+
# regular text output
|
|
98
|
+
$ letta -p "hi there"
|
|
99
|
+
Hi! How can I help you today?
|
|
100
|
+
|
|
101
|
+
# structured output
|
|
102
|
+
$ letta -p "hi there" --output-format json
|
|
103
|
+
{
|
|
104
|
+
"type": "result",
|
|
105
|
+
"subtype": "success",
|
|
106
|
+
"is_error": false,
|
|
107
|
+
"duration_ms": 5454,
|
|
108
|
+
"duration_api_ms": 2098,
|
|
109
|
+
"num_turns": 1,
|
|
110
|
+
"result": "Hi! How can I help you today?",
|
|
111
|
+
"session_id": "agent-8ab431ca-63e0-4ca1-ba83-b64d66d95a0f",
|
|
112
|
+
"usage": {
|
|
113
|
+
"input_tokens": 294,
|
|
114
|
+
"output_tokens": 97
|
|
115
|
+
}
|
|
116
|
+
}
|
|
117
|
+
```
|
|
118
|
+
|
|
60
119
|
### Permissions
|
|
61
120
|
|
|
62
121
|
**Tool selection** (controls which tools are loaded):
|
|
@@ -102,8 +161,8 @@ First, install Bun if you don't have it yet: [https://bun.com/docs/installation]
|
|
|
102
161
|
bun install
|
|
103
162
|
|
|
104
163
|
# run the CLI from TypeScript sources (pick up changes immediately)
|
|
105
|
-
bun run dev
|
|
106
|
-
bun run dev
|
|
164
|
+
bun run dev
|
|
165
|
+
bun run dev -- -p "Hello world" # example with args
|
|
107
166
|
```
|
|
108
167
|
|
|
109
168
|
### Build + link the standalone binary
|