autonomous-coding-toolkit 1.0.1 → 1.0.3
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/README.md +7 -8
- package/package.json +19 -5
package/README.md
CHANGED
|
@@ -4,11 +4,9 @@
|
|
|
4
4
|
|
|
5
5
|
# Autonomous Coding Toolkit
|
|
6
6
|
|
|
7
|
-
|
|
8
|
-
|
|
9
|
-
**A learning system for autonomous AI coding.** Fresh context per batch, quality gates between every step, 79 community lessons that prevent the same bug twice, and telemetry that makes the system smarter with every run.
|
|
7
|
+
An autonomous AI coding agent that executes implementation plans with quality gates, fresh context per batch, and 79 community-contributed lessons that prevent the same bug twice. Built for [Claude Code](https://docs.anthropic.com/en/docs/claude-code) — works as a Claude Code plugin (interactive) or npm CLI for headless CI/CD execution.
|
|
10
8
|
|
|
11
|
-
|
|
9
|
+
> **Goal:** Code better than a human on large projects — not by being smarter on any single batch, but by compounding learning across thousands of batches across hundreds of users.
|
|
12
10
|
|
|
13
11
|
## Install
|
|
14
12
|
|
|
@@ -155,10 +153,11 @@ Core skill chain forked from [superpowers](https://github.com/obra/superpowers)
|
|
|
155
153
|
|
|
156
154
|
The toolkit's design is grounded in peer-reviewed research. Key papers:
|
|
157
155
|
|
|
158
|
-
- **SWE-bench Pro** (Xia et al., 2025) — 1,865 programming problems; removing specifications degraded agent success from 25.9% to 8.4%
|
|
159
|
-
- **
|
|
160
|
-
- **Lost in the Middle** (Liu et al., Stanford TACL 2024) — Information placed mid-context suffers up to 20 percentage point accuracy loss
|
|
161
|
-
- **OOPSLA 2025
|
|
156
|
+
- [**SWE-bench Pro**](https://arxiv.org/pdf/2509.16941) (Xia et al., 2025) — 1,865 programming problems; removing specifications degraded agent success from 25.9% to 8.4%
|
|
157
|
+
- [**Context Rot**](https://research.trychroma.com/context-rot) (Hong et al., Chroma 2025) — 11 of 12 models scored below 50% of short-context performance at 32K tokens
|
|
158
|
+
- [**Lost in the Middle**](https://arxiv.org/abs/2307.03172) (Liu et al., Stanford TACL 2024) — Information placed mid-context suffers up to 20 percentage point accuracy loss
|
|
159
|
+
- [**Agentic Property-Based Testing**](https://arxiv.org/html/2510.09907v1) (OOPSLA 2025) — Property-based testing finds ~50x more mutations per test than traditional unit tests
|
|
160
|
+
- [**Bugs in LLM-Generated Code**](https://arxiv.org/abs/2403.08937) (Tambon et al., 2024) — Empirical taxonomy of AI code generation failures
|
|
162
161
|
- **Cooper Stage-Gate** — Projects with stable, upfront definitions are 3x more likely to succeed
|
|
163
162
|
|
|
164
163
|
16 research reports synthesizing 25+ papers: [`docs/RESEARCH.md`](docs/RESEARCH.md)
|
package/package.json
CHANGED
|
@@ -1,11 +1,17 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "autonomous-coding-toolkit",
|
|
3
|
-
"version": "1.0.
|
|
4
|
-
"description": "Autonomous AI coding pipeline
|
|
3
|
+
"version": "1.0.3",
|
|
4
|
+
"description": "Autonomous AI coding agent pipeline with quality gates, fresh context per batch, 79 community lessons, and compounding learning for Claude Code",
|
|
5
5
|
"license": "MIT",
|
|
6
6
|
"author": "Justin McFarland <parthalon025@gmail.com>",
|
|
7
7
|
"homepage": "https://github.com/parthalon025/autonomous-coding-toolkit",
|
|
8
|
-
"repository":
|
|
8
|
+
"repository": {
|
|
9
|
+
"type": "git",
|
|
10
|
+
"url": "https://github.com/parthalon025/autonomous-coding-toolkit.git"
|
|
11
|
+
},
|
|
12
|
+
"bugs": {
|
|
13
|
+
"url": "https://github.com/parthalon025/autonomous-coding-toolkit/issues"
|
|
14
|
+
},
|
|
9
15
|
"bin": {
|
|
10
16
|
"act": "./bin/act.js"
|
|
11
17
|
},
|
|
@@ -35,13 +41,21 @@
|
|
|
35
41
|
"keywords": [
|
|
36
42
|
"autonomous-coding",
|
|
37
43
|
"ai-agents",
|
|
38
|
-
"
|
|
44
|
+
"ai-coding",
|
|
45
|
+
"agentic-ai",
|
|
46
|
+
"code-generation",
|
|
47
|
+
"claude",
|
|
39
48
|
"claude-code",
|
|
49
|
+
"quality-gates",
|
|
40
50
|
"tdd",
|
|
41
51
|
"lessons-learned",
|
|
42
52
|
"headless",
|
|
53
|
+
"ci-cd",
|
|
54
|
+
"automation",
|
|
43
55
|
"multi-armed-bandit",
|
|
44
56
|
"code-review",
|
|
45
|
-
"pipeline"
|
|
57
|
+
"pipeline",
|
|
58
|
+
"testing",
|
|
59
|
+
"developer-tools"
|
|
46
60
|
]
|
|
47
61
|
}
|