Rewrite Fable skill and onboarding docs

[REX-codebase]

Summary

• replace the dense skill mandate list with a usable Think -> Prove -> Attack -> Write workflow
• add a practical first-session guide with real MCP request examples and troubleshooting
• correct the README quick start and pinned Cursor version
• remove a stale hard-coded test count from the contributor guide

Verification

• 14 focused manifest, wheel, and install-skill tests pass
• wheel build and install-skill end-to-end check pass
• root and packaged skill files are byte-identical
• Markdown links, fences, and git diff --check pass

The full 490-test baseline still has four environment-related failures that predate and do not touch these documentation-only changes.

Summary by CodeRabbit

• Documentation

  • Added a Quick Start workflow covering MCP installation, optional Agent Skill setup, task initiation, evidence tracking, and execution safeguards.
  • Expanded practical guidance for installing and using Fable Mode, including first-session steps, requirements, troubleshooting, and configuration.
  • Updated Fable Mode references to describe the THINK → PROVE → ATTACK → WRITE lifecycle, timed execution controls, testing, adversarial review, and completion reporting.
  • Added links to practical and Agent Skill documentation.

• Tests

  • Added a dependency-free smoke-test command using Python’s unittest discovery.
  • Retained guidance to include tests with new behavior.

[coderabbitai]

Caution

Review failed

The pull request is closed.

Warning

.coderabbit.yaml has a parsing error

The CodeRabbit configuration file in this repository has a parsing error and default settings were used instead. Please fix the error(s) in the configuration file. You can initialize chat with CodeRabbit to get help with the configuration file.

Parsing errors (1)

Validation error: Too big: expected string to have <=250 characters at "tone_instructions"

⚙️ Configuration instructions

• Please see the configuration documentation for more information.
• You can also validate your configuration using the online YAML validator.
• If your editor has YAML language server enabled, you can add the path at the top of this file to enable auto-completion and validation: # yaml-language-server: $schema=https://coderabbit.ai/integrations/schema.v2.json

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Advanced

Run ID: 85b559a2-74b0-4186-8f96-e84f0403345a

📥 Commits

Reviewing files that changed from the base of the PR and between 5eb73c2 and 3b86a87.

📒 Files selected for processing (5)

• CONTRIBUTING.md
• README.md
• SKILL.md
• docs/stop-ai-agents-writing-too-early.md
• skills/fable-mode/SKILL.md

---

📝 Walkthrough

Walkthrough

The pull request replaces architecture-heavy Fable Mode guidance with an evidence-gated workflow. It updates the README, practical guide, skill specifications, and contributor testing instructions.

Changes

Fable Mode documentation

Layer / File(s)	Summary
Evidence-gated workflow specification SKILL.md, skills/fable-mode/SKILL.md	The skill specifications define the THINK, PROVE, ATTACK, and WRITE lifecycle, evidence and invariant requirements, timed unlocking, implementation checks, completion records, action mappings, references, and failure handling.
Installation and first-session guide docs/stop-ai-agents-writing-too-early.md	The guide documents MCP and optional Agent Skill installation, first-session steps, gate requirements, boundaries, troubleshooting, and current references.
Quick start and testing guidance README.md, CONTRIBUTING.md	The README adds Quick start instructions and documentation links. CONTRIBUTING.md adds a dependency-free unittest smoke command and removes the fixed test count.

Estimated code review effort: 2 (Simple) | ~10 minutes

Change: Other

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch REX-codebase-patch-1

---

_{Comment @coderabbitai help to get the list of available commands.}