Skip to content

Commit 8bf2d50

Browse files
committed
docs: align README with CompText-Sparkctl branding
1 parent 5a62fd2 commit 8bf2d50

1 file changed

Lines changed: 126 additions & 111 deletions

File tree

README.md

Lines changed: 126 additions & 111 deletions
Original file line numberDiff line numberDiff line change
@@ -1,58 +1,59 @@
1-
# sparkctl
1+
# CompText-Sparkctl
22

33
<p align="center">
4-
<img src="assets/branding/sparkctl-logo.png" alt="sparkctl logo" width="520">
4+
<img src="assets/branding/sparkctl-logo.png" alt="CompText-Sparkctl logo" width="520">
55
</p>
66

7-
## Antigravity-CompText v7 / SPARK Context Layer
8-
97
<div align="center">
108

11-
[![GitHub Stars](https://img.shields.io/github/stars/ProfRandom92/Antigravity-Comptextv7?style=for-the-badge&color=yellow)](https://github.com/ProfRandom92/Antigravity-Comptextv7/stargazers)
9+
[![GitHub Stars](https://img.shields.io/github/stars/ProfRandom92/comptext-sparkctl?style=for-the-badge&color=yellow)](https://github.com/ProfRandom92/comptext-sparkctl/stargazers)
1210
[![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg?style=for-the-badge)](https://opensource.org/licenses/MIT)
13-
[![Python](https://img.shields.io/badge/Python-3.10+-3776AB.svg?style=for-the-badge&logo=python&logoColor=white)](https://www.python.org/)
14-
[![Rust](https://img.shields.io/badge/Rust-integrated-000000.svg?style=for-the-badge&logo=rust&logoColor=white)](https://www.rust-lang.org/)
15-
[![Security](https://img.shields.io/badge/Security-SHA--256%20Sidecar-red.svg?style=for-the-badge)](#-security-model)
16-
[![SPARK](https://img.shields.io/badge/SPARK-Hackathon%20Track-2b6cb0.svg?style=for-the-badge)](#-spark-hackathon-track)
11+
[![Rust](https://img.shields.io/badge/Rust-CLI-000000.svg?style=for-the-badge&logo=rust&logoColor=white)](https://www.rust-lang.org/)
12+
[![Security](https://img.shields.io/badge/Security-SHA--256%20Sidecar-red.svg?style=for-the-badge)](#-safety-boundaries--claim-hygiene)
13+
[![SPARK](https://img.shields.io/badge/SPARK-Context%20Artifacts-2b6cb0.svg?style=for-the-badge)](#-spark-context-artifacts)
1714

18-
**Deterministic trace compression and operational context validation for autonomous agent systems.**
15+
**Deterministic Rust CLI for CompText trace packaging, replay-sidecar validation, and SPARK-ready context artifacts.**
1916

2017
</div>
2118

2219
---
2320

24-
## 🚀 Overview
21+
## Overview
22+
23+
**CompText-Sparkctl** is a local Rust command-line toolkit for turning agent or operational traces into compact, verifiable packages and SPARK-style context artifacts.
24+
25+
The project is built around one hard rule: **compression must not destroy replay-critical state**. CompText separates compact linguistic payloads from replay-sensitive metadata, then validates the result with a replay sidecar, SHA-256 integrity anchors, schema checks, and offline context validation flows.
26+
27+
The repository currently exposes two CLI entry points:
2528

26-
`sparkctl` is the unified operations controller and command-line interface for the **Antigravity-CompText v7 / SPARK Context Layer**. It streamlines local diagnostics, codebase validation, pipeline lifecycle orchestration, and integration demonstrations under a clean, secure, and offline execution environment.
29+
- `sparkctl` — the validated compatibility surface for local diagnostics, Rust validation, context pipeline checks, demo execution, and handoff checks.
30+
- `agy-ct` — the newer CompText command surface, currently wired to safe compatibility wrappers for selected existing `sparkctl` functionality.
31+
32+
Previously, this project used the working name **Antigravity-CompText v7**. The current public project name is **CompText-Sparkctl**.
33+
34+
---
2735

28-
CompText v7 separates compressible linguistic payloads from replay-critical state, then reconstructs canonical traces with cryptographic sidecar integrity. The SPARK Context Layer provides offline Rust pipelines for packaging, schema validation, token-light rendering, and structured context checking.
36+
## What it does
2937

30-
Classic lossy trace compression fails when validators expect exact tool order, commitment tokens, state hashes, and canonical replay strings. CompText v7 avoids that failure mode by splitting each trace into two coordinated streams:
38+
CompText-Sparkctl validates the split between compressible trace content and replay-critical state:
3139

3240
| Layer | Purpose | Target property |
3341
|---|---|---|
3442
| **CompText payload** | Pruned, compact linguistic trace | Lower token and transport cost |
3543
| **Replay sidecar** | Tool sequence, commitments, hashes, state anchors | Deterministic reconstruction in the validated scope |
3644
| **SHA-256 audit chain** | Integrity metadata over critical replay data | Tamper-sensitive validation |
3745
| **Holdout validator** | Non-adaptive replay verification | Stable replay score in benchmark runs |
46+
| **SPARK context artifacts** | Structured operational context and rendered summaries | Local validation and handoff readiness |
3847

39-
---
40-
41-
## 🛠 Command Surface
42-
43-
`sparkctl` consolidates all operations under a single command-line interface:
44-
45-
* **`sparkctl doctor`**: Diagnoses local project readiness by verifying the presence of key configuration files, JSON schemas, and output artifacts.
46-
* **`sparkctl rust-validate`**: Automates execution of local crate quality checks (`cargo fmt`, `cargo check`, `cargo test`, `cargo clippy`).
47-
* **`sparkctl context-all`**: Coordinates the local context pipeline sequence (`context-build`, `context-render`, `context-validate`).
48-
* **`sparkctl spark-demo`**: Orchestrates the full end-to-end integration lifecycle (`compress`, `context-build`, `context-render`, `context-validate`).
49-
* **`sparkctl handoff-check`**: Validates file completeness and command availability to ensure clean repository handoff status.
48+
Classic lossy compression fails when validators expect exact tool order, commitment tokens, state hashes, and canonical replay strings. CompText-Sparkctl keeps those replay-sensitive fields outside the lossy zone.
5049

5150
---
5251

53-
## ⚡ Quickstart
52+
## Command Surface
53+
54+
### `sparkctl`
5455

55-
To run `sparkctl`, navigate to the `agy7rust` directory and use `cargo run`:
56+
`sparkctl` is the validated operations controller:
5657

5758
```bash
5859
cd agy7rust
@@ -63,60 +64,79 @@ cargo run --bin sparkctl -- spark-demo
6364
cargo run --bin sparkctl -- handoff-check
6465
```
6566

66-
### Artifact Outputs
67-
The SPARK pipeline generates and validates the following key artifacts under `artifacts/spark/`:
68-
- `artifacts/spark/extraction.spkg`: The deterministically compressed trace package containing payload and sidecar metadata.
69-
- `artifacts/spark/context.json`: The compiled structured operational context mapping dependencies and security parameters.
70-
- `artifacts/spark/context_render.txt`: The token-light rendered text view for verification and summarization.
67+
Validated command responsibilities:
7168

72-
---
69+
- `sparkctl doctor` checks local project readiness.
70+
- `sparkctl rust-validate` runs local Rust quality checks.
71+
- `sparkctl context-all` runs the local context build/render/validate sequence.
72+
- `sparkctl spark-demo` runs the local end-to-end demonstration flow.
73+
- `sparkctl handoff-check` checks local repository handoff readiness.
7374

74-
## 🔒 Safety, Boundaries & Claim Hygiene
75+
### `agy-ct`
7576

76-
* **Offline Execution:** All subcommands operate completely locally. Offline behavior was deterministic in the validated test scope.
77-
* **Leak Boundaries:** Configured leak checks passed in the validated scope.
78-
* **Local Handoff Checks:** `sparkctl handoff-check` validates local repository readiness and file availability only; it does not verify remote CI or GitHub Actions status.
79-
* **Compatibility:** No official SPARK compatibility claim is made.
80-
* **Compliance:** No compliance claim (such as EU AI Act compliance) is made.
81-
* **Risks:** No blocking risks found in the validated scope.
77+
`agy-ct` is the newer command surface for CompText-Sparkctl. It currently provides the command tree and safe compatibility wrappers without introducing a new run orchestrator.
8278

83-
---
79+
```bash
80+
cd agy7rust
81+
cargo run --bin agy-ct -- --help
82+
cargo run --bin agy-ct -- doctor
83+
cargo run --bin agy-ct -- validate
84+
cargo run --bin agy-ct -- handoff
85+
cargo run --bin agy-ct -- demo
86+
cargo run --bin agy-ct -- context all
87+
```
8488

85-
## 📈 Project Phase Status
89+
Current wrapper mapping:
8690

87-
* **Phase 3 (Context Layer):** Build, Render, and Validate slices — **Complete**
88-
* **Phase 4 (sparkctl Command Surface):** `doctor`, `rust-validate`, `context-all`, `spark-demo`, and `handoff-check` subcommands — **Complete**
89-
* **Phase 5 (Release & Branding):** Release documentation & branding integration — **In Progress**
91+
| `agy-ct` command | Existing validated backend |
92+
|---|---|
93+
| `agy-ct doctor` | `sparkctl::doctor::run_doctor()` |
94+
| `agy-ct validate` | `sparkctl::rust_validate::run_rust_validate()` |
95+
| `agy-ct handoff` | `sparkctl::handoff_check::run_handoff_check()` |
96+
| `agy-ct demo` | `sparkctl::spark_demo::run_spark_demo()` |
97+
| `agy-ct context all` | `sparkctl::context_all::run_context_all()` |
98+
99+
Other `agy-ct` commands remain explicit placeholders until their implementation phase is approved.
90100

91101
---
92102

93-
## 🏛 Architecture
103+
## SPARK Context Artifacts
104+
105+
The local SPARK-style demo and context pipeline generates and validates artifacts under `artifacts/spark/`:
106+
107+
- `artifacts/spark/extraction.spkg` — compact trace package containing payload and replay-sidecar metadata.
108+
- `artifacts/spark/context.json` — structured operational context for validation and handoff.
109+
- `artifacts/spark/context_render.txt` — token-light rendered context view for review and summarization.
94110

95-
CompText v7 is built around one hard rule: **payload compression must never destroy replay-critical state**.
111+
These artifacts are intended for local, reproducible validation and review workflows.
112+
113+
---
114+
115+
## Architecture
96116

97117
```mermaid
98118
flowchart TD
99-
raw["Raw Agent / LMCache Trace"]
119+
raw["Raw Agent / Operational Trace"]
100120
101-
subgraph split["KVTC Split Layer"]
121+
subgraph split["CompText Split Layer"]
102122
splitter["Trace Splitter"]
103123
classifier["Replay-Critical Field Classifier"]
104124
end
105125
106126
subgraph payload["Compressible Payload Stream"]
107127
text["Linguistic Payload"]
108128
prune["Redundancy Pruning"]
109-
compact["CompText v7 Compact Payload"]
129+
compact["Compact Payload"]
110130
end
111131
112-
subgraph sidecar["Deterministic Replay Sidecar"]
132+
subgraph sidecar["Replay Sidecar"]
113133
order["Tool Order"]
114134
commits["Commitment Tokens"]
115135
state["State Hashes"]
116136
anchors["Validation Anchors"]
117137
end
118138
119-
subgraph context["Operational Context Layer"]
139+
subgraph context["SPARK Context Layer"]
120140
ctxbuild["context-build"]
121141
ctxrender["context-render"]
122142
ctxvalidate["context-validate"]
@@ -166,7 +186,7 @@ flowchart LR
166186
H["Context validation anchors"]
167187
end
168188
169-
subgraph output["CompText v7 Package"]
189+
subgraph output["CompText Package"]
170190
I["Compact payload"]
171191
J["Replay sidecar"]
172192
K["SHA-256 audit chain"]
@@ -186,48 +206,41 @@ flowchart LR
186206

187207
---
188208

189-
## 🦀 Rust Integration
209+
## Rust Integration
190210

191-
Rust is integrated as the execution path for the parts that should be fast, deterministic in the validated scope, and easy to audit:
211+
Rust is the hardened local execution path for components that need to be fast, auditable, and deterministic in the validated scope:
192212

193213
- byte-level payload handling
194214
- deterministic hashing and verification
195215
- replay-sidecar validation
196216
- schema-sidecar validation
197217
- operational context build/render/validate flows
198-
- low-overhead execution inside local validation workflows
218+
- local CLI validation and handoff checks
199219

200-
Python remains useful as the reference and experimentation layer. Rust is the direction for hardened execution.
220+
Python remains useful as a reference and experimentation layer. Rust is the direction for hardened execution.
201221

202222
---
203223

204-
## 🔒 Security Model
205-
206-
CompText v7 does not treat compression as a purely cosmetic optimization. Every replay-sensitive field is part of the integrity surface.
207-
208-
The sidecar protects:
209-
210-
- tool execution order
211-
- commitment and control tokens
212-
- final state hash
213-
- replay metadata
214-
- validation-critical anchors
215-
- context validation anchors
224+
## Safety, Boundaries & Claim Hygiene
216225

217-
If a compressed package is modified without updating the expected integrity chain, reconstruction fails loudly instead of producing a misleading replay.
226+
- **Offline execution:** Validated commands operate locally. Offline behavior was deterministic in the validated test scope.
227+
- **Leak boundaries:** Configured leak checks passed in the validated scope.
228+
- **Local handoff checks:** `sparkctl handoff-check` checks local repository readiness and file availability only; it does not verify remote CI or GitHub Actions status.
229+
- **SPARK:** No official SPARK compatibility claim is made.
230+
- **Compliance:** No compliance claim, including EU AI Act compliance, is made.
231+
- **Risk statement:** No blocking risks found in the validated scope.
218232

219233
Non-claims:
220234

221-
- no official SPARK compatibility claim
222-
- no EU AI Act compliance claim
223235
- no legal evidentiary-status claim
224236
- no forensic certainty claim
225-
- no MCP server capability
237+
- no MCP server capability claim
226238
- no RAG, embeddings, vector database, or external tool-orchestration layer
239+
- no production-readiness, certification, or compliance claim
227240

228241
---
229242

230-
## 📊 Benchmarks
243+
## Benchmarks
231244

232245
Current validation targets are based on the existing CompText v7 benchmark profile:
233246

@@ -238,19 +251,20 @@ Current validation targets are based on the existing CompText v7 benchmark profi
238251
| C | Regex pruning | ~68 % of raw | 1.00 | No forensic integrity |
239252
| D/E | Blind reduction | variable | 0.0 on complex traces | Loses temporal/state-critical tokens |
240253

241-
The design goal is not maximum textual compression at any cost. The goal is maximum safe reduction under strict deterministic replay constraints.
254+
The design goal is not maximum textual compression at any cost. The goal is maximum safe reduction under strict replay constraints.
242255

243256
---
244257

245-
## 📦 Repository Map
258+
## Repository Map
246259

247260
```text
248261
.
249262
├── .agent/ # Local agent skills used for gated implementation
250-
├── .antigravitycli/ # Antigravity CLI/runtime configuration
263+
├── .antigravitycli/ # Local agent runtime configuration
251264
├── Comptextv7/ # CompText v7 integration surface
252-
├── agy7rust/ # Rust CLI path for SPARK-style packaging and context flows
253-
├── artifacts/spark/ # Generated SPARK demo/package/context artifacts
265+
├── agy7rust/ # Rust CLI path for packaging, validation, and context flows
266+
├── artifacts/spark/ # Generated SPARK-style demo/package/context artifacts
267+
├── assets/branding/ # Project branding assets
254268
├── benchmarks/ # Benchmark profiles and comparison material
255269
├── core/ # KVTC / replay core components
256270
├── datasets/ # Fixtures and trace datasets
@@ -263,34 +277,18 @@ The design goal is not maximum textual compression at any cost. The goal is maxi
263277

264278
---
265279

266-
## 🤝 Contributing
280+
## Project Phase Status
267281

268-
Contributions are welcome. The project is especially interested in work that improves determinism, compression quality, auditability, Rust hardening, or SPARK-style administrative AI verification.
269-
270-
Good first contribution areas:
271-
272-
- add new trace fixtures
273-
- add SPARK-style extraction fixtures
274-
- improve benchmark coverage
275-
- document edge cases
276-
- add Rust-side validation tests
277-
- tighten schema-sidecar checks
278-
- improve CI reproducibility
279-
- extend operational context validation while preserving leak boundaries
280-
281-
Contribution flow:
282-
283-
1. Fork the repository.
284-
2. Create a feature branch: `git checkout -b feature/your-improvement`.
285-
3. Make a focused change.
286-
4. Run tests locally.
287-
5. Open a pull request with a clear before/after explanation.
288-
289-
Please keep PRs small, reproducible, and validation-oriented.
282+
- **Phase 3:** Operational context layer — complete.
283+
- **Phase 4:** `sparkctl` command surface — complete.
284+
- **Phase 5:** Release README and branding integration — complete.
285+
- **Phase 6A:** `agy-ct` CLI architecture handbook — complete.
286+
- **Phase 6B:** `agy-ct` binary and command tree — complete.
287+
- **Phase 6C:** `agy-ct` compatibility wrappers — complete in the validated command scope.
290288

291289
---
292290

293-
## 🛣 Roadmap
291+
## Roadmap
294292

295293
- [x] Deterministic replay-sidecar architecture
296294
- [x] SHA-256 integrity anchoring
@@ -301,28 +299,45 @@ Please keep PRs small, reproducible, and validation-oriented.
301299
- [x] Offline SPARK demo fixtures
302300
- [x] SPARK operational context model
303301
- [x] SPARK context build/render/validate CLI flow
304-
- [x] Unified CLI `sparkctl` (`doctor`, `rust-validate`, `context-all`, `spark-demo`, `handoff-check`)
305-
- [ ] CI benchmark snapshots
302+
- [x] Unified `sparkctl` CLI
303+
- [x] `agy-ct` command surface
304+
- [x] `agy-ct` compatibility wrappers
305+
- [ ] `agy-ct run` orchestrator
306+
- [ ] JSON report export
307+
- [ ] Notebook bundle/export
308+
- [ ] Context compiler roadmap
309+
- [ ] Agent-control event model
306310
- [ ] Fresh-clone GitHub verification workflow
307-
- [ ] Public examples for custom trace datasets
308-
- [ ] v8 generalization layer for enterprise agent pipelines
309311

310312
---
311313

312-
## 🌟 Support the project
314+
## Contributing
315+
316+
Contributions are welcome, especially work that improves determinism, compression quality, auditability, Rust hardening, SPARK-style administrative AI verification, or reproducible validation.
317+
318+
Good first contribution areas:
319+
320+
- add new trace fixtures
321+
- add SPARK-style extraction fixtures
322+
- improve benchmark coverage
323+
- document edge cases
324+
- add Rust-side validation tests
325+
- tighten schema-sidecar checks
326+
- improve CI reproducibility
327+
- extend operational context validation while preserving leak boundaries
313328

314-
If this project helps you reason about safer agent traces, compression, or deterministic replay, consider leaving a star. It makes the project easier to discover and helps attract contributors who care about reliable agent infrastructure.
329+
Please keep pull requests small, reproducible, and validation-oriented.
315330

316331
---
317332

318-
## 📄 License
333+
## License
319334

320335
This project is released under the MIT License.
321336

322337
---
323338

324339
<div align="center">
325340

326-
**CompText v7: compress the noise, preserve the proof.**
341+
**CompText-Sparkctl: compress the noise, preserve the proof.**
327342

328343
</div>

0 commit comments

Comments
 (0)