Fable Domain
Generates a sourced domain workflow: adapter, flowchart, trap fixture, eval
Test report
- Verdict
- Tested · Works
- Score
- Tested
- Jul 16, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · 8a45cd5
Task: assemble a fable-method domain adapter for the new 'localization / translation QA' sector — first without skill, then strictly according to the skill's body. Baseline run (I copied the schema from an existing marketing.md) yielded 39 lines, 6/8 mandatory TEMPLATE.md sections, zero sources, and no flowcharts. Skill run yielded 74 lines, 8/8 sections, a mermaid flowchart, and 3 sources with access dates, of which I actually verified CLDR plural rules via fetch and confirmed six categories and minimal pairs; ISO 17100 confirmed the requirement for bilingual revision by a second linguist. The repository's own CI (python .github/checks.py) passed green on both adapters — it only greps two headers, so the real difference was provided by the skill itself, not CI. Honesty spot-check: README statements (Haiku 2→6, Sonnet 9→10, Opus 8→9) matched committed eval/results/round12-13 to the point, including zeros unfavorable to the author. Minus: the repository's install.sh does not install fable-domain at all (0 mentions), and the skill's body refers to references/domains/TEMPLATE.md, which is located in the fable-method skill — hence two `cp` lines in install; I did not run the trap-fixture and smoke-eval from Stage 3-4, and this was not checked.
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 4/5
- Triggers reliably 4/5
- Output vs. baseline 8/10
- Docs & honesty 5/5
What Fable Domain does
Turns a sector into a Fable-method domain adapter: a step-by-step workflow with a mermaid flowchart, a fraud table, a trap fixture, and a Sources section where every named regulation carries a fetched link and access date. Triggers on "/fable-domain <sector>", "make a skill for <domain>", or adding a domain to the fable method. Refuses outright on licensure and high-harm domains and stops early when a sector does not differ from the coding default.
How to install Fable Domain
git clone https://github.com/Sahir619/fable-method
mkdir -p ~/.claude/skills
cp -r fable-method/skills/fable-method ~/.claude/skills/fable-method
cp -r fable-method/skills/fable-domain ~/.claude/skills/fable-domain
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger Fable Domain
-
/fable-domainGenerates a sourced domain workflow: adapter, flowchart, trap fixture, eval
It also activates on plain-language prompts like these:
-
Make me a workflow adapter for localization QA with sourced regulations -
Add a new domain to the fable method for translation quality checks -
Build a fraud table and trap fixture for this new industry workflow
Frequently asked questions
- Is the Fable Domain skill free?
- Yes. The skill itself is free from Sahir619/fable-method. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Fable Domain work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 16, 2026. Verdict: Tested · Works. Task: assemble a fable-method domain adapter for the new 'localization / translation QA' sector — first without skill, then strictly according to the skill's body. Baseline run (I copied the schema from an existing marketing.md) yielded 39 lines, 6/8 mandatory TEMPLATE.md sections, zero sources, and no flowcharts. Skill run yielded 74 lines, 8/8 sections, a mermaid flowchart, and 3 sources with access dates, of which I actually verified CLDR plural rules via fetch and confirmed six categories and minimal pairs; ISO 17100 confirmed the requirement for bilingual revision by a second linguist. The repository's own CI (python .github/checks.py) passed green on both adapters — it only greps two headers, so the real difference was provided by the skill itself, not CI. Honesty spot-check: README statements (Haiku 2→6, Sonnet 9→10, Opus 8→9) matched committed eval/results/round12-13 to the point, including zeros unfavorable to the author. Minus: the repository's install.sh does not install fable-domain at all (0 mentions), and the skill's body refers to references/domains/TEMPLATE.md, which is located in the fable-method skill — hence two `cp` lines in install; I did not run the trap-fixture and smoke-eval from Stage 3-4, and this was not checked.
- What is the Fable Domain SkillProof Score?
- 8.4/10 — installs cleanly 4/5, triggers reliably 4/5, output vs. baseline 8/10, docs & honesty 5/5.
- How do I install Fable Domain?
- Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
- Can I use Fable Domain with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
- The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.