From tofu-harness
Use at the start of any non-trivial task (build / fix / analyze / deliver) to run it under the tofable working discipline — honest decomposition, prompt refinement via /prompt when installed (basic-prompting fallback when not), evidence-first execution, and honest completion. Also use when asked to "apply tofable", "work like fable", or "tofable로 진행".
How this skill is triggered — by the user, by Claude, or both
Slash command
/tofu-harness:tofableThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Run the task through 5 checkpoints. Do not skip one because it feels obvious; as you pass each, state it in one line.
Run the task through 5 checkpoints. Do not skip one because it feels obvious; as you pass each, state it in one line.
/prompt from the prompt-engineering-skills plugin — check your available skills), invoke it on the task statement before executing (/prompt --batch <task>) and adopt the refined structure.hooks/ in the tofable repo): the gates bounce you when evidence is missing at stop time; this skill makes that evidence exist before the gate ever fires.hooks/README.md (FABLE_GATE_OFF=1).Guides completion of development work by verifying tests, detecting environment, and presenting structured options for merge, PR, or cleanup.
Guides creation and editing of skills using test-driven development with pressure scenarios and subagents to verify agent compliance.
Dispatches multiple subagents concurrently for independent tasks without shared state. Use when facing 2+ unrelated failures or subsystems that can be investigated in parallel.
npx claudepluginhub treylom/tofu-harness --plugin tofu-harness