neurodock

Motor-aware pacing — decomposes a goal, shows each estimate padded by the profile buffer, names why.

Motor-aware pacing — decomposes a goal, shows each estimate padded by the profile buffer, names why.

Morning brief — what changed overnight, what matters today, one next-action per project.

Transcript → four-section brief (my asks, others' asks, decisions, ambiguous items with verbatim quotes). Records decisions into the cognitive graph.

/resume command — reconstruct yesterday's mental state from the cognitive graph, without inventing facts.

Reformats responses to Answer-First when sessions run long; surfaces the user's own prior intent if past their break threshold. Also activates on design-critique contexts (Figma reviews, comp reviews, prototype walk-throughs, spec reviews) so the verdict lands before the reasoning.

Surfaces prior decision evidence on repeat-validation requests; declines further re-analysis until new information appears.

Converts prose, meeting notes, or overwhelm-dumps into Mermaid diagrams (flowchart / sequence / mindmap).

Walk through cutting a PyPI release of one of the five MCP servers — bump version, write CHANGELOG, run tests and evals, tag, publish.

Detect whether the user's current prompt is a semantic repeat of recent prompts within a rolling window. Returns an advisory signal; never blocks.

Close the currently open chronometric session, optionally attaching a verbatim summary. Triggers the auto-summary side effect used by weekly_rollup.

Open a chronometric session by calling mark_session_start with the user's stated intent. Returns a session_id that later interventions quote back verbatim.

Break a vague goal into 3–12 atomic 5–90 minute tasks with acceptance criteria and dependencies via the NeuroDock task-fractionator. The acceptance bar is "I can start now".

Persist a subject-predicate-object fact in the local cognitive graph. Load this BEFORE calling record_fact to send the correct input shape on the first attempt.

Decode subtext and ambiguity in an incoming corporate message via the NeuroDock translation server. Returns explicit ask, ranked subtext, ambiguity spans anchored verbatim, and a recommended next action.

Transcript → four-section brief (my asks, others' asks, decisions, ambiguous items with verbatim quotes). Records decisions into the cognitive graph.

Check new CSS / JSX for animation that runs by default — the project default is `motion: reduced` and any motion must gate on prefers-reduced-motion.

Grep changed markdown / mdx / tsx string literals on the current branch for banned terms — superpower, spectrum colours, clinical phrasing, marketing intensifiers.

Verify that mcp-translation prompts reference field names that match the schemas. Catches the class of bug where a prompt asks for `confidence` but the schema requires `certainty`.

For any change to mcp-translation or its eval corpus, verify every fixture satisfies ambiguous_items[].quoted_span.text == transcript[start_char:end_char]. The schema invariant from ADR 0005.

Detect whether the user's current prompt is a semantic repeat of recent prompts within a rolling window. Returns an advisory signal; never blocks.

Close the currently open chronometric session, optionally attaching a verbatim summary. Triggers the auto-summary side effect used by weekly_rollup.

Open a chronometric session by calling mark_session_start with the user's stated intent. Returns a session_id that later interventions quote back verbatim.

Break a vague goal into 3–12 atomic 5–90 minute tasks with acceptance criteria and dependencies via mcp-task-fractionator. The acceptance bar is "I can start now".

Scaffold a fresh ~/.neurodock/profile.yaml from one of the ND-aware presets under profiles/. Use on first install or when starting a new device.

Persist a subject-predicate-object fact in the local cognitive graph. Load this BEFORE calling record_fact to avoid the documented six-retry validation loop.

Walk through cutting a browser-extension release — bump manifest, build for Chrome / Firefox / Edge, zip artefacts, optionally run E2E, submit to stores.

Walk through adding a new tool to one of the six MCP servers under packages/mcp-*. Touches schema, handler, registration, tests, ADR delta, CHANGELOG.

Decode subtext and ambiguity in an incoming corporate message via mcp-translation. Returns explicit ask, ranked subtext, ambiguity spans anchored verbatim, and a recommended next action.

Validate a profile.yaml against the canonical schema and surface structured errors. Run before editing the profile by hand, before sync, and before reporting a profile-related bug.

Assess a diff for version-bump implications, missing CHANGELOG entries, and stale docs. Run before opening a PR.

A 25-minute Pomodoro coach — opens a session anchored to a stated intent, surfaces a break suggestion at the user-chosen duration, and closes the session on user signal.

Three civil-servant flows — draft a structured briefing note, synthesise stakeholder consultation responses, or set meeting-marathon chronometric thresholds for a sitting-day.

Three engineering-manager flows — prep a 1:1 from prior context, decompose a sprint review, or walk a blameless postmortem.

Three lawyer/paralegal flows — brief a matter from prior context, run a forward-window deadline check, or translate an incoming email.

Three PM flows — render a stakeholder map for a feature, audit decisions on a project in temporal order, or shape a verdict-first exec summary within a word budget.

Three researcher flows — start a literature review by decomposing it into atomic reading and synthesis tasks, record a paper and its authors as graph facts, or recall the papers an author has written.

Three software-engineer IC flows — prep a code review from prior reviewer comments and decisions, plan a deep-work block of N hours, or write an async standup from recorded facts.

Three long-form-writing flows — plan a writing block as scene-shaped atomic tasks, recall what the graph remembers about a character or source, or check whether you are ruminating on a chapter.

Reformats responses to Answer-First when sessions run long; surfaces the user's own prior intent if past their break threshold. Also activates on design-critique contexts (Figma reviews, comp reviews, prototype walk-throughs, spec reviews) so the verdict lands before the reasoning.