菜单
Before each significant action, check whether it serves the current goal. After 3 adjacent off-goal actions, name the drift explicitly.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Manually route the current request through the installed neurodivergent-support skill pack. Use only when the user explicitly invokes `/nd` in Claude Code, `$nd` in Codex, selects ND Router from the skills UI, asks to activate ND support for the current task, or requests ND status/off controls. Select the smallest relevant subset of sibling `nd-*` skills and continue useful work instead of merely describing them.
When presenting options or a backlog, show only the top 3. Hidden items are noted but not shown until one choice is completed.
Before starting a new task thread, surface any tasks that are more than halfway complete. Ask whether to close one first. One-tap dismiss. Never blocks.
Detect complex or multi-step requests and suggest structured planning before acting. Prevents diving into implementation before the problem is understood.
Stabilize task execution when the user explicitly reports frustration, shame, panic, anger, rejection, or overwhelm that is affecting decisions. Preserve work, separate observable facts from interpretations, lower immediate decision load, and choose one reversible next action without turning the exchange into therapy.
Redesign the task environment when a known intention repeatedly fails at the point of action. Use when reminders arrive too early or late, tools or state are hard to find, choices or steps create friction, a system depends on memory, or repeated failure shows that effort alone is not enough.
| name | nd-inhibit |
| description | Before each significant action, check whether it serves the current goal. After 3 adjacent off-goal actions, name the drift explicitly. |
Inhibitory control: the executive function that stops you from acting on impulse when the action doesn't serve the current goal. In coding, this shows up as scope creep, yak-shaving, and "while I'm here I'll also fix..." patterns that expand a 1-hour task into a 3-day refactor.
The inhibitory control skill is a lightweight, non-blocking pre-action check: does this action serve the goal we stated at the start of this task?
Apply this skill during any task that has a defined goal. Run the check:
Do NOT apply to: Read-only operations (reading files, searching), clearly pre-planned steps, or when the user has explicitly expanded scope.
State the current goal (from the start of the task): "Active goal: [goal]"
Before a potentially off-goal action, ask silently: Does [this specific action] directly serve [the current goal], operate on the intended target, and respect every explicit "must not" constraint?
If yes: proceed without comment.
If no, stop the action and return to the constraint ledger. A plausible improvement is still off-goal when it changes the wrong artifact, deletes something the user protected, or violates an earlier correction.
If uncertain: Emit a one-line annotation before the action:
"→ [action] — this serves [goal] because [reason]"
After 3 consecutive actions that seem tangentially related to the goal: Name the drift:
"⚠ Drift check: the last 3 actions were [list]. The original goal was [goal]. Want to continue this direction, or return to the original path?"
Never block correct in-scope execution. Do not ask the user to reconfirm work they already authorized. Pause only for a real constraint conflict, new authority, or material ambiguity that inspection cannot resolve.
Inhibitory control failure in ADHD often manifests as off-task intrusions — thoughts or impulses that pull attention away from the current task. In coding, the trigger is often seeing something that could be "improved while I'm here." The brain treats this as an opportunity (reward signal) rather than a distraction.
The 3-action threshold before naming drift is deliberate: 1 tangential action is normal problem-solving, 2 is still plausibly connected, 3 establishes a pattern that warrants naming. Naming the drift without blocking it gives the human the choice to re-engage the goal — which is the correct intervention, since sometimes the drift IS the right thing.
