메뉴
Interrupt research spirals after N turns without concrete output. Surface the pattern, name the original goal, and offer a one-question redirect to execution.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
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-research-gate |
| description | Interrupt research spirals after N turns without concrete output. Surface the pattern, name the original goal, and offer a one-question redirect to execution. |
A well-documented executive function pattern: discovery and analysis feel rewarding, so the brain stays in exploration mode past the point of diminishing returns. Each new finding reinforces continuing. Without an explicit interrupt, a session can produce 10 turns of research and zero shipped artifacts.
This is inhibitory control failure — the inability to stop a rewarding behavior when a goal requires switching to a less immediately rewarding one (execution).
Apply this skill when you notice you have done 8 or more consecutive turns that are primarily:
Without any turn that produced:
Reading, searching, listing files, browsing, and writing another analysis document do not reset the counter merely because a tool or shell command ran.
Count and name the pattern. Before the next research action, surface it explicitly:
"I've done [N] consecutive research turns since the last concrete output. The stated goal was: [original goal]."
Offer the single redirect question. Do not continue researching. Ask exactly:
"Want to pick one finding and build it now — or continue exploring? (I'll follow your call.)"
On 'build it now': If one finding is already the clear recommendation, implement it immediately. Ask the user to choose only when multiple materially different findings remain and the choice cannot be inferred from their stated goal.
On 'keep exploring': Reset the counter mentally and continue — but surface it again after another N turns.
Counter: Default threshold is 8 turns. This should feel like a check-in, not a gate. Never block.
Executive dysfunction research (Rabinovici et al., 2015; PMC4455841) identifies inhibition failure as a core EF deficit: difficulty stopping a cognitive pattern even when it no longer serves the goal. Treatment: environmental manipulation — make the pattern visible so the user can choose, rather than continuing automatically.
The one-question format is deliberate. Presenting multiple options (keep going, or build this, or build that, or pivot, or...) recreates the same choice paralysis the skill is designed to interrupt.
