Skip to content

CRP Scribe

Future product

CRP Scribe is a roadmap product. The implementation code is not in this repository today, the managed SaaS at scribe.crprotocol.io is not live, and no pricing tier is available for purchase. The documentation below describes the planned surface once it is built. Self-hosted deployment is the likely first delivery model.

Generate long-form documents of any length with your own LLM keys.

CRP Scribe removes the output wall. Books, manuals, regulatory reports, and whitepapers are generated chapter-by-chapter using CRP's continuation engine, fact graph, and quality scoring - while keeping your data and API keys under your control.

Any Model, Any Length

CRP Scribe will use the continuation engine to generate documents of unlimited length - books, manuals, whitepapers, regulatory reports. The output wall doesn't exist.

Product Identity
Product CRP Scribe
Type Long-form document generation engine (roadmap)
Powered By Context Relay Protocol™ - continuation engine, knowledge graph, voice profiling, quality scoring
Deployment Self-hosted first; managed SaaS at scribe.crprotocol.io planned later
License Elastic License 2.0 (ELv2)

How It Works

You provide: Topic + Template + LLM API Key
                    │
                    ▼
┌─────────────────────────────────────────────────┐
│              CRP Scribe (Planned)               │
│                                                 │
│  ┌─────────────┐  ┌──────────────────────────┐  │
│  │ Generation   │  │ CRP Subsystems:          │  │
│  │ Orchestrator │──│  • Extraction            │  │
│  │ (chapter by  │  │  • Knowledge Graph (CKF) │  │
│  │  chapter)    │  │  • Voice Profiling       │  │
│  └─────┬───────┘  │  • Quality Scoring       │  │
│        │          │  • Continuation Engine   │  │
│  ┌─────┴─────┐    │  • Circuit Breaker       │  │
│  │ Dashboard  │    │  • Fact Integrity Chain  │  │
│  │ • Progress │    │  • Cross-Chapter Valid.  │  │
│  │ • Templates│    │  • PII/Injection Scan    │  │
│  │ • Documents│    │                          │  │
│  └────────────┘    └──────────────────────────┘  │
│                                                 │
└──────────────────────┬──────────────────────────┘
                       │  Calls your LLM per chapter
                       ▼
              Your LLM Provider
              (OpenAI / Anthropic / Ollama / LM Studio)

You get: 50K+ word document with per-chapter quality metrics and full fact graph.

Overview

Every LLM has a finite output window. Ask for a 30-chapter book and you get 8 chapters before truncation. CRP Scribe eliminates this limitation entirely.

It combines CRP's continuation engine with a template system and outline generator to produce complete, coherent, multi-chapter documents in a single command.

Quick Start

Not available yet

Sign-up, dashboard, and API endpoints at scribe.crprotocol.io are not live.

1. Sign Up

Planned: visit scribe.crprotocol.io and create your account.

2. Generate Documents

Planned: use the web dashboard to generate unlimited-length documents. Choose a template, provide your topic, and let the CRP continuation engine handle the rest.

3. API Access (Pro & Enterprise)

Planned API example:

import requests

response = requests.post(
    "https://scribe.crprotocol.io/api/v1/generate",
    headers={"X-API-Key": "crc_your_scribe_key"},
    json={
        "topic": "AI Governance and the EU AI Act",
        "template": "whitepaper",
        "num_chapters": 10,
    },
)

result = response.json()
print(f"Chapters: {len(result['chapters'])}")

Templates

CRP Scribe is planned to ship with built-in templates:

Template Best For Default Chapters
whitepaper Industry analysis, thought leadership, regulatory briefs 8
technical_manual API docs, deployment guides, runbooks 12
guide How-to guides, tutorials, onboarding docs 8
report Audit reports, assessments, findings 6
book Full-length books, comprehensive references 15

Outline Generation

Planned: generate a structured outline before writing. The outline can be LLM-powered (when a provider is available) or use intelligent defaults.

Full Document Generation

Planned: the dashboard will provide real-time progress tracking as each chapter is generated. Documents will be auto-saved and available for download in Markdown format.

Output Format

Planned output:

  • Full text - complete document as a single string
  • Per-chapter breakdown - each chapter with title, content, word count, and CRP quality tier (S/A/B/C/D)
  • Metadata - generation timestamp, model used, template applied
  • Export - download as Markdown, or via API as JSON

Pricing

Not final

Pricing tiers below are illustrative; they are not live products.

  • Free

    • 50 generations/mo
    • 2 templates (whitepaper, guide)
    • 1 LLM provider (OpenAI)
    • Max 5,000 words per document
    • Markdown export
    • 7-day version history

    $0/mo (planned)

  • Pro

    Everything in Free, plus:

    • 500 generations/mo
    • All templates
    • All LLM providers
    • Max 50,000 words per document
    • MD, PDF, DOCX export
    • 90-day version history
    • API access
    • Email support

    $39/mo (planned)

  • Enterprise

    Everything in Pro, plus:

    • Unlimited generations
    • Custom templates
    • Private models (Ollama, LM Studio)
    • Unlimited document length
    • All export formats + API export
    • Unlimited version history
    • SSO (SAML / OIDC)
    • Priority support

    Custom (planned)

Provider Support

CRP Scribe will work with any CRP-supported LLM provider:

Provider Setup
OpenAI export OPENAI_API_KEY=sk-...
Anthropic export ANTHROPIC_API_KEY=sk-ant-...
Ollama --provider ollama --model llama3
LM Studio --provider lmstudio --model local-model

Contact