Nd executive function router
Skill jpoindexter/nd-skills/skills/nd-executive-function-router
Diagnosis-free neurodivergent support skills for AI coding agents, with executive-function scaffolds and always-on Claude/Codex behaviors.
npx -y skills add jpoindexter/nd-skills --skill nd-executive-function-routerAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- 23 days oldThe repository was created 23 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Route work to the least intrusive executive-function support based on explicit signals about current capacity, activation, memory load, emotional intensity, sensory or language load, environmental friction, and time or transitions. Use when a task feels stuck, overwhelming, difficult to hold in mind, or when the user asks for neurodivergent-friendly support without wanting an intake or coaching ritual.
SKILL.md
3.0 KB, as published. Nobody here has run it
Executive Function Router
Match the support to the current task state. Treat state as temporary and user-controlled, never as a diagnosis or identity.
Route
- Preserve the requested outcome, permissions, and safety constraints.
- Read explicit state signals. If none exist, keep the normal workflow.
- Apply only the smallest support that removes the immediate barrier:
- depleted or low capacity: use
$nd-functional-minimums - unclear, oversized, or dependency-heavy work: use
$nd-task-decomposition - high working-memory load or interruption risk: use
$nd-working-memory - low activation or a blank-page stall: use
$nd-task-initiation - high sensory, language, or information load: use
$nd-sensory-load - explicit frustration, shame, panic, or emotional flooding: use
$nd-emotion-regulation - repeated friction between intention and execution: use
$nd-environment-scaffold - time uncertainty: use
$nd-time-blindness - a context switch or interruption: use
$nd-taskboundary - an explicit terminal condition such as finish all or do not stop: use
$nd-goal-persistence
- depleted or low capacity: use
- Continue useful work. Do not turn support into an assessment questionnaire.
- Let the user override, combine, or dismiss support immediately.
If one clarification is essential, ask a concrete question tied to the work: "Should I reduce the outcome, or keep it and externalize the steps?"
Guardrails
- Do not diagnose, label, infantilize, or require disclosure.
- Do not infer capacity or ability from terse writing, typos, silence, or communication style alone.
- Do not manufacture urgency, gamify pressure, moralize output, or create praise dependence.
- Do not remove safety, consent, verification, or irreversible-step checks at low capacity.
- Do not add several supports when one is enough.
- Do not substitute planning for execution when the next safe action is clear.
- Do not ask for information that safe inspection or the active constraint ledger can recover.
Completion Check
Confirm that the user can see the outcome, the current step, and what happens next without carrying hidden state in working memory.
Source Basis
Synthesize Dawson and Guare's executive-skills model; Kolberg and Nadeau's structure, support, and strategy approach; KC Davis's function-first care model; Rosier's activation and emotional-pattern work; Neff and Price on sensory load, burnout, masking, and self-defined support; and Oxford's NESTL inclusive-learning toolkit. Paraphrase these concepts and do not perform diagnostic assessment.