nd-skills

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.

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.

Reduce work to the smallest safe and useful outcome when the user explicitly reports low capacity, depletion, burnout, overwhelm, or asks for the minimum viable version. Preserve safety and function while deferring polish, and execute the reduced outcome when permissions allow.

Preserve an explicit terminal goal across subtasks and keep executing safe, authorized work until its end-to-end done condition is observed or a genuine blocker is reached. Use when the user says finish all, do not stop, continue until done, complete every item, or supplies a multi-item backlog whose intermediate milestones must not be mistaken for completion.

Detect and name hyperfocus sessions that have gone deep on a rabbit hole. Surface the original goal and offer a structured way back.

Before each significant action, check whether it serves the current goal. After 3 adjacent off-goal actions, name the drift explicitly.

Interrupt research spirals after N turns without concrete output. Surface the pattern, name the original goal, and offer a one-question redirect to execution.

Before destructive or hard-to-reverse operations, pause to verify the action matches the stated goal. A one-sentence pre-flight check, not a gate.

Reduce language, information, sensory, and uncertainty load while preserving complete information. Use when the user requests direct or scannable communication, explicitly reports overload or low capacity, is navigating an unfamiliar flow, or needs clear expectations and examples.

When stuck on the same approach for 3+ iterations, detect the pattern and propose an alternative strategy. Read prior failure logs before attempting a complex fix.

Turn a vague, oversized, dependency-heavy, or initiation-resistant outcome into independently verifiable units with a visible first action and observable definition of done. Use when the user cannot identify where to begin, hidden dependencies block action, or a complex request needs execution structure after planning is accepted.

When a user faces a blank page, explicitly says they are stuck or avoiding a known task, or does not know where to start, create the smallest concrete first action and, when needed, add one truthful interest, novelty, challenge, feedback, or body-doubling bridge. Never invent urgency or use shame.

Mark explicit cognitive task boundaries when topics shift or an interruption risks losing state. Save the active outcome, last verified fact, current action, blocker, and exact re-entry step before beginning the new context.

Externalize uncertain time with calibrated best, realistic, and worst ranges, hidden costs, checkpoints, and stopping rules. Use for estimates, open-ended research or refinement, work likely to expand, and tasks where elapsed-time feedback should revise the forecast.

Track and surface the ratio of ideas captured vs items shipped. Warn when the capture:ship ratio is high — the ideas-rich/finish-poor pattern made visible.

Maintain explicit task context across subtasks and persist it in one durable working surface when interruptions, multiple turns, or multiple agents could lose the thread. Re-state active goals when context grows long, track what is in flight, and provide an exact re-entry action.