Fulcrum 7 of 18
Deciding how much care the work deserves
At the start of any task, before implementation.
What goes wrong here
Every task gets the same effort. A change to payments or login is treated like a typo fix, or a trivial change gets a full ceremony. Nobody remembered to ask for the deeper checks.
How it compounds if nothing catches it
High-stakes work that was under-checked is where real users get hurt, and where the person loses the ability to trust the agent unattended. Trivial work that was over-checked spends the person's attention on approvals that didn't need them.
What the harness does here
The governer is the step that scores how much rigor a task deserves, from 0 to 100, by how much a mistake could cost and whom it would touch. It routes the task on that score. Low scores proceed. Higher scores get, in turn, a written plan, an independent evaluator and human review before anything reaches users. A gate (pre-tool-governer-gate.sh) blocks edits until the session has a score. After three blocks it softens to a reminder, so it can't trap the agent in a loop. At the start of a task the governer also writes verification contracts: the specific runnable checks that will prove the task done, which the stop gate enforces later. governer2 is an experimental version that turns the skills a task is predicted to need into steps that can't be skipped. The prediction comes from predict-required-skills2, which learns from what agents actually did in past sessions. task-orchestrator locks in a required sequence of steps.
The pieces that act here
- /governer
A skill: an instruction file an agent loads by name.
- /governer2
Opt-in upgrade to /governer: same scoring and routing core, plus a notebook built from your own past sessions that can be asked which steps this kind of task has needed before, forced workflow stages the agent can't quietly skip, and full observability (every oracle query and answer printed, every skipped stage logged). Off by default. 'governer2' is worth keeping as a distinct opt-in name rather than folding into /governer until its oracle has actually been exercised enough to judge.
- pre-tool-governer-gate.sh
A hook: a script Claude Code runs on its own at a set moment.
- /verification-contracts
Task-specific verification contracts — written before work starts, recorded with `alignment-harness contract add`, and enforced by the Stop hook at task end. The full reference for the per-message contract prompt.
- /predict-required-skills2
Predicts the skill sequence a new scope of work needs, grounded in why past sessions actually reached for each skill — not a guess. Works two ways: a local fallback that reads your own past session transcripts (no setup), or an optional NotebookLM oracle built from an automatically-growing corpus of your own \"why I used this skill\" annotations. Invoke at the start of any distinct scope to get the skill sequence, in order, grounded in cited precedent.
- /task-orchestrator
Automatic task orchestration — ensures every agent task gets the right pipeline steps (/align, /governer, /decompose, /declare-scope, /reflect, /consume, /complete-agentic-task) without the person having to ask. Predicts which steps are needed based on task characteristics, prints them, and spawns guardian agents.