fable-ready
Write-mode readiness audit for Claude Fable 5. Scans CLAUDE.md + skills + configs for the five documented breakage families, patches what you approve, stamps a dated FABLE-READY verdict. $49 once.
unzip fable-ready-v1.0.1.zip -d ~/.claude/skills/ # delivered instantly after purchaseAbout
Fable 5 changed the rules underneath Opus-era setups. Prescriptive CLAUDE.md files degrade output. "Show your reasoning" prompts end in refusals that return HTTP 200 and look like success. Old configs carry params the API now rejects.
The five rule families: reasoning-display triggers (refusal risk), prescriptive step-lists (quality), missing output discipline (cost), missing boundaries (surprises), mechanics drift (rejected params, stale model strings, timeout assumptions). Every rule in rules.md carries an evidence grade and a source. Diffed against Anthropic's own migration guide and Fable fallback cookbook.
The split: a deterministic stdlib scanner catches what regexes can catch — reproducibly. The model judges what regexes can't. Patches are approval-gated; meaning-changing patches are never auto-applied. Ends in a dated FABLE-READY verdict; re-run after patching and the verdict updates.
Also inside: effort-map.md (task type → effort level, with the cost math), long-run.md (verifier subagents, anti-fabrication), memory-bootstrap.md (the lesson-file system).
Built, validated, and audited via Ship Kit on the Solid State production line — Fable 5 patched its own migration. The full run is published as a case study. Rules dated 2026-06; v1.x updates included.
Pre-publish audit for ClawHub skills. Nine checks, one verdict: READY or FIX FIRST. Pass the security scan on the first upload.
Find and verify real research papers through a keyless index — no login, no API key. Search by topic or method, map a field's citation graph, pull the in-body passages that prove a claim, or scan GitHub for engineering prior art. Grounds every answer in papers the index actually returns — never invents titles or arXiv IDs.
A system, not a pile of skills. Six skills + orchestrator + shared project memory: validate, position, audit (write-mode), launch. $99 once.