@hilbras/omninode 0.5.2 → 0.7.0
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 +48 -4
- package/dist/cli/index.js +990 -15
- package/dist/cli/index.js.map +1 -1
- package/dist/index.d.ts +546 -61
- package/dist/index.js +815 -10
- package/dist/index.js.map +1 -1
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -18,7 +18,7 @@ existing intelligence; it does not try to become another model.
|
|
|
18
18
|
|
|
19
19
|
## Status
|
|
20
20
|
|
|
21
|
-
**v0.
|
|
21
|
+
**v0.7.0 — Phase 6: Multi-AI Report System** (see [docs/DEVELOPMENT_PLAN.md](docs/DEVELOPMENT_PLAN.md)
|
|
22
22
|
for the full plan and roadmap).
|
|
23
23
|
|
|
24
24
|
What exists today:
|
|
@@ -53,6 +53,27 @@ What exists today:
|
|
|
53
53
|
results (summary, reports, errors, timestamps)
|
|
54
54
|
- `TaskStore` interface with a local-first JSON implementation in
|
|
55
55
|
`.omninode/tasks.json` (atomic writes; swap in any backend)
|
|
56
|
+
- **Pipeline engine** (`src/pipelines/`):
|
|
57
|
+
- Pipelines defined in `omninode.yaml` under `project.pipelines`, validated
|
|
58
|
+
before every run (duplicate ids, unknown dependencies, cycles, missing
|
|
59
|
+
agents)
|
|
60
|
+
- Sequential execution by definition order; explicit `depends_on` unlocks
|
|
61
|
+
parallel branches, and research steps fan out to several agents in
|
|
62
|
+
parallel (§15)
|
|
63
|
+
- Step kinds: `research` (multi-agent fan-out), `collect`, `analyze`,
|
|
64
|
+
`plan` (calls a model as `provider:model-id` via the chat layer),
|
|
65
|
+
`execute`/`custom` (hand-off to an execution agent)
|
|
66
|
+
- Failure handling: retries per step, `on-success`/`on-failure`/`always`
|
|
67
|
+
conditions, combined context threaded downstream; every task runs through
|
|
68
|
+
the TaskEngine and every run is persisted to `.omninode/pipelines.json`
|
|
69
|
+
- **Multi-AI report system** (`src/reports/`):
|
|
70
|
+
- Report collection from finished tasks — structured protocol reports pass
|
|
71
|
+
through; plain text output is normalized via deterministic extraction
|
|
72
|
+
(findings/recommendations/evidence sections, severity guessing)
|
|
73
|
+
- Persistence in `.omninode/reports.json` with full source attribution
|
|
74
|
+
- Aggregation (§17): duplicate/similar findings merged via deterministic
|
|
75
|
+
token-set similarity, sources preserved, highest severity kept
|
|
76
|
+
- Combined reports generated automatically at the end of every pipeline run
|
|
56
77
|
- Typed error hierarchy (`src/errors/`)
|
|
57
78
|
- Leveled logger with pluggable sink (`src/logger/`)
|
|
58
79
|
- Configuration system: `omninode.yaml` with `${ENV_VAR}` expansion and strict
|
|
@@ -111,6 +132,30 @@ omninode task status task-xxxxxx
|
|
|
111
132
|
omninode task list
|
|
112
133
|
```
|
|
113
134
|
|
|
135
|
+
Orchestrate several AI systems with a pipeline (§14):
|
|
136
|
+
|
|
137
|
+
```yaml
|
|
138
|
+
project:
|
|
139
|
+
pipelines:
|
|
140
|
+
- id: repo-audit
|
|
141
|
+
steps:
|
|
142
|
+
- id: research
|
|
143
|
+
kind: research
|
|
144
|
+
agents: [kimi, gemini, qwen]
|
|
145
|
+
- id: plan
|
|
146
|
+
kind: plan
|
|
147
|
+
model: my-gateway:chatgpt
|
|
148
|
+
- id: execute
|
|
149
|
+
kind: execute
|
|
150
|
+
agent: opencode
|
|
151
|
+
```
|
|
152
|
+
|
|
153
|
+
```bash
|
|
154
|
+
omninode pipeline run repo-audit "Audit the authentication module"
|
|
155
|
+
omninode report combined # the aggregated intelligence of that run
|
|
156
|
+
omninode report show combined-xxx # grouped findings with per-source attribution
|
|
157
|
+
```
|
|
158
|
+
|
|
114
159
|
See [omninode.yaml.example](omninode.yaml.example) for a full configuration.
|
|
115
160
|
|
|
116
161
|
### Configuration
|
|
@@ -144,9 +189,8 @@ Phases 0–11 are laid out in [docs/DEVELOPMENT_PLAN.md](docs/DEVELOPMENT_PLAN.m
|
|
|
144
189
|
| v0.2.0 | Provider System |
|
|
145
190
|
| v0.3.0 | OmniHilbras Integration (optional, like every provider) |
|
|
146
191
|
| v0.4.0 | Agent System |
|
|
147
|
-
|
|
|
148
|
-
| v0.
|
|
149
|
-
| v0.7.0 | Multi-AI Reports |
|
|
192
|
+
| v0.5.0 | Roles & Tasks |
|
|
193
|
+
| **v0.7.0** | **Multi-AI Reports (current)** |
|
|
150
194
|
| v0.8.0 | Remembera (optional memory integration) |
|
|
151
195
|
| v0.9.0 | Planner |
|
|
152
196
|
| v0.10.0 | End-to-End Runtime |
|