v1.0.0 - Initial Release

The Fable Method v1.0.0

A structured problem-solving discipline loop that gives any model Fable 5 grade thinking.

Four Hermes Agent skills:

• fable-method (think): the 7-step loop - classify, define done, gather evidence, decide, act surgically, verify by observation, report honestly
• fable-loop (act): orchestrated version with parallel evidence subagents via delegate_task and adversarial verifier agents
• fable-judge (prove): adversarial verification of finished work - re-runs every claimed check, diffs what changed, hunts frauds, verdicts VERIFIED / CAVEATS / REFUTED
• fable-domain (grow): domain adapter generator - discusses use case, researches from real sources, generates adapter + trap fixture + smoke eval

Eight domain adapters:

marketing, research, data analysis, business/ops, finance, legal/compliance, design/UX, devops/infrastructure

Three eval trap scenarios with real A/B results:

• discount-calc (spec-vs-test conflict)
• inventory-value (assessment trap - editing is the wrong move)
• config-deploy (unauthorized action trap - docs are not authorization)

Install:

hermes plugins install ardhaecosystem/fable-method --enable
bash ~/.hermes/plugins/fable-method/install.sh

Key features:

• Forced artifacts at decision points (INTENT, AUTH, TWINS, PENDING lines)
• Hard bounds (3 failed verify cycles = stop, 2 fruitless lookups = stop searching)
• Authorization gate (documentation is not authorization)
• Twin check (search the whole project for copies of a bug)
• Artifact gate (terminal sweep adds any owed-and-missing report lines)
• 18 failure modes mapped to the step that prevents each
• Eight decision flowcharts as executable pseudocode