@nathapp/nax 0.61.2 → 0.62.0-canary.1
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 +6 -3
- package/dist/nax.js +33298 -32037
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -15,6 +15,7 @@ nax is an **orchestrator, not an agent** — it doesn't write code itself. It dr
|
|
|
15
15
|
- **Extensible** — plugin system for routing, review, reporting, and post-run actions
|
|
16
16
|
- **Language-aware** — auto-detects Go, Rust, Python, TypeScript from manifest files; adapts commands, test structure, and mocking patterns per language
|
|
17
17
|
- **Semantic review** — LLM-based behavioral review against story acceptance criteria; catches stubs, placeholders, and out-of-scope changes
|
|
18
|
+
- **Adversarial review** — LLM-based adversarial code review that probes for input handling, error paths, and abandoned implementations
|
|
18
19
|
|
|
19
20
|
## Install
|
|
20
21
|
|
|
@@ -46,7 +47,7 @@ See [docs/](docs/) for full guides on configuration, test strategies, monorepo s
|
|
|
46
47
|
## How It Works
|
|
47
48
|
|
|
48
49
|
```
|
|
49
|
-
(plan →) acceptance setup → route → execute → verify → review → escalate → loop → regression gate → acceptance
|
|
50
|
+
(plan →) acceptance setup → route → execute → verify → review (semantic + adversarial) → escalate → loop → regression gate → acceptance
|
|
50
51
|
```
|
|
51
52
|
|
|
52
53
|
1. **Plan** *(optional)* — Generate `prd.json` from a spec file using an LLM
|
|
@@ -55,7 +56,7 @@ See [docs/](docs/) for full guides on configuration, test strategies, monorepo s
|
|
|
55
56
|
4. **Context** — Gather relevant code, tests, and project standards per story
|
|
56
57
|
5. **Execute** — Run agent session (Claude Code, Codex, Gemini CLI, or ACP)
|
|
57
58
|
6. **Verify** — Run scoped tests; rectify on failure before escalating
|
|
58
|
-
7. **Review** — Run lint + typecheck; autofix before escalating
|
|
59
|
+
7. **Review** — Run lint + typecheck + semantic review + adversarial review; autofix before escalating
|
|
59
60
|
8. **Escalate** — On repeated failure, retry with a higher model tier
|
|
60
61
|
9. **Loop** — Repeat steps 3–8 per story until all pass or a cost/iteration limit is hit
|
|
61
62
|
10. **Regression gate** — Run full test suite after all stories pass
|
|
@@ -142,10 +143,12 @@ After all stories pass, nax runs the full test suite once. If it fails, it retri
|
|
|
142
143
|
|
|
143
144
|
See [Regression Gate Guide](docs/guides/regression-gate.md).
|
|
144
145
|
|
|
145
|
-
### Parallel Execution
|
|
146
|
+
### Parallel & Isolated Execution
|
|
146
147
|
|
|
147
148
|
Stories are batched by compatibility (same model tier, similar complexity) and run in parallel within each batch. Use `--parallel <n>` to control concurrency. Sequential mode uses a deferred regression gate; parallel mode always runs regression at the end.
|
|
148
149
|
|
|
150
|
+
Even in sequential mode, stories can be isolated in per-story git worktrees (`execution.storyIsolation: "worktree"`) to prevent cross-story state leakage.
|
|
151
|
+
|
|
149
152
|
See [Parallel Execution Guide](docs/guides/parallel-execution.md).
|
|
150
153
|
|
|
151
154
|
### Monorepo Support
|