razzou-harness 0.1.2 → 0.1.3
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/CHANGELOG.md +6 -0
- package/README.md +11 -7
- package/package.json +9 -2
package/CHANGELOG.md
CHANGED
|
@@ -3,6 +3,12 @@
|
|
|
3
3
|
The notable changes in each release of razzou-harness. Before 1.0, a new minor version
|
|
4
4
|
may change how a project is set up, and its notes say how.
|
|
5
5
|
|
|
6
|
+
## 0.1.3 (2026-10-11)
|
|
7
|
+
|
|
8
|
+
- More npm keywords, for a team of agents in sandboxes.
|
|
9
|
+
- The README says what a module is, a part of the system you want to build, and how to
|
|
10
|
+
size one: its design fits in the Architect's context.
|
|
11
|
+
|
|
6
12
|
## 0.1.2 (2026-10-11)
|
|
7
13
|
|
|
8
14
|
- The source is public, on GitHub, and the package links to it.
|
package/README.md
CHANGED
|
@@ -2,11 +2,11 @@
|
|
|
2
2
|
|
|
3
3
|
**Your AI dev team, with you in charge.**
|
|
4
4
|
|
|
5
|
-
You describe a module
|
|
6
|
-
at a time, and a Manager reviews each one.
|
|
7
|
-
[OpenShell](https://github.com/NVIDIA/OpenShell) sandbox
|
|
8
|
-
feature is checked in a fresh one, and nothing reaches `main`
|
|
9
|
-
request. A small daemon runs the loop, following Anthropic's
|
|
5
|
+
You describe a module, a part of the system you want to build. An Architect designs it
|
|
6
|
+
with you, Developers build it one feature at a time, and a Manager reviews each one.
|
|
7
|
+
Every agent works in its own [OpenShell](https://github.com/NVIDIA/OpenShell) sandbox
|
|
8
|
+
with no credentials, every feature is checked in a fresh one, and nothing reaches `main`
|
|
9
|
+
until you merge its pull request. A small daemon runs the loop, following Anthropic's
|
|
10
10
|
[Effective harnesses for long-running agents](https://www.anthropic.com/engineering/effective-harnesses-for-long-running-agents).
|
|
11
11
|
|
|
12
12
|
- [How it works](#how-it-works)
|
|
@@ -141,8 +141,12 @@ plans it finds there (see [Configuration](#configuration)).
|
|
|
141
141
|
|
|
142
142
|
## Concepts
|
|
143
143
|
|
|
144
|
-
- **Module:** a
|
|
145
|
-
plans. Modules are built one after another.
|
|
144
|
+
- **Module:** a part of the system you want to build, with one Architect and one
|
|
145
|
+
accepted proposal, then its plans. Modules are built one after another. Size a module
|
|
146
|
+
so that its design fits in one conversation. With the default model, Opus 5.5, the
|
|
147
|
+
Architect's context holds up to a million tokens, and a well-sized module never fills
|
|
148
|
+
it. If a module's design keeps growing, split it into smaller modules: each gets its
|
|
149
|
+
own Architect, with a fresh context.
|
|
146
150
|
- **Proposal:** the Architect's design document, in versions v1, v2 and so on. You
|
|
147
151
|
accept one.
|
|
148
152
|
- **Plan:** features that ship together as one pull request. A module's plans are built
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "razzou-harness",
|
|
3
|
-
"version": "0.1.
|
|
3
|
+
"version": "0.1.3",
|
|
4
4
|
"description": "Your AI dev team, with you in charge: sandboxed Claude Code agents design, build and review your project, and you merge.",
|
|
5
5
|
"license": "MIT",
|
|
6
6
|
"author": "razzou",
|
|
@@ -15,11 +15,18 @@
|
|
|
15
15
|
"keywords": [
|
|
16
16
|
"claude",
|
|
17
17
|
"claude-code",
|
|
18
|
+
"anthropic",
|
|
18
19
|
"ai-agents",
|
|
19
20
|
"coding-agents",
|
|
21
|
+
"agentic-coding",
|
|
22
|
+
"multi-agent",
|
|
23
|
+
"agent-team",
|
|
24
|
+
"agent-orchestration",
|
|
20
25
|
"agent-harness",
|
|
21
|
-
"openshell",
|
|
22
26
|
"sandbox",
|
|
27
|
+
"sandboxed",
|
|
28
|
+
"agent-sandbox",
|
|
29
|
+
"openshell",
|
|
23
30
|
"terraform",
|
|
24
31
|
"aws",
|
|
25
32
|
"github-app"
|