@mindrian_os/cli 2.0.0-beta.29 → 2.0.0-beta.31

package/skills/funding/SKILL.md

@@ -0,0 +1,110 @@
+---
+name: funding
+description: Track grant opportunities through their lifecycle
+license: BSL-1.1. See LICENSE for complete terms (Business Source License 1.1, Change Date 2030-04-16 to Apache License 2.0).
+help_jtbd: "See the grants and funding paths matching your room."
+argument-hint: "[list|add|update]"
+body_shape: B (Semantic Tree)
+hitl_shape: "F.8"
+hitl_why: "Grant-lifecycle candidates are surfaced as an independent set the navigator triages in any order."
+# Phase 267.3-06, ruled in 267.3-CLASSIFICATION.md (Row 13, navigator-ruled NEEDS-RULING): first delivery at commands/funding.md:57, a stage/deadline/staleness tracking report of the navigator's own pipeline, not the Tavily-sourced discovery flow the rule doc's worked example names.
+interactive_first_reward: "--none (diagnostic surface)"
+serves_jtbd: ["prepare-pitch", "decide-pursue"]
+teaching: "When you need to track grants and investors as a real pipeline, /mos:funding manages the lifecycle from spotted to applied to decided. Funding is a process, not a one-shot event."
+allowed-tools: Read Write Bash Glob AskUserQuestion
+# --- Phase 143.3 connector frontmatter ---
+connector:
+  connects_to_spine: true
+  sensor_triggers: []
+  reach_id: context_block
+  sub_mode: funding
+  framework: null
+  posture: hold
+  hierarchy_rank: 37
+  filing: none
+  plan_gated: false
+  web_scope: null
+---
+
+<!-- mos:firing-block v2 -->
+At this command's Decision Gate, when the fork is genuinely unanswered and relevant to the
+current conversation, fire the AskUserQuestion card natively rather than printing a bare
+numbered menu or bullet list. Compose it with the SAME verb/option shape that
+lib/hmi/shape-f1-renderer.cjs (renderShapeF1) produces and that lib/hmi/selector-dispatcher.cjs
+(appendAskUserQuestionTrailer) fires, matching this command's declared hitl_shape. Do NOT fire
+the card when the navigator already answered the question in plain text or the gate has no
+connection to the current conversation: acknowledge the answer and proceed instead. Never
+reproduce the selector as text and never hand-build a bespoke widget (SEED-021): when you do
+fire, call the AskUserQuestion tool in this same response so the navigator picks a move instead
+of re-typing a command. Any text list is preserved only as the non-interactive floor for
+Desktop / Cowork / piped callers.
+<!-- /mos:firing-block -->
+
+# /mos:funding
+
+> Track grant opportunities through the funding lifecycle. Promote discoveries from opportunity-bank, advance through stages, and monitor your funding pipeline.
+
+## Subcommands
+
+### `list`
+
+Show funding pipeline: all entries grouped by stage with deadlines and days-since-update.
+
+**Example:** "Show me my funding pipeline" or `/mos:funding list`
+
+Larry displays entries organized by stage (Discovered > Researched > Applying > Submitted), highlights upcoming deadlines, and flags stale entries that haven't been updated in 14+ days.
+
+### `create [opportunity-slug]`
+
+Promote an opportunity from opportunity-bank to the funding pipeline. Creates a per-opportunity folder at `room/funding/{slug}/` with initial stage **Discovered**.
+
+**Example:** "Start tracking the NSF SBIR grant" or `/mos:funding create nsf-sbir-phase1`
+
+This creates:
+- `STATUS.md` with stage: discovered, wikilink to source opportunity
+- `metadata.yaml` with structured data from the opportunity artifact
+
+The funding entry cross-references its source via `[[opportunity-bank/{source}]]` wikilink, creating a graph edge back to the discovery.
+
+### `advance [slug]`
+
+Move a funding entry to the next stage. Larry confirms the transition and asks for a note. Stage order is enforced:
+
+```
+Discovered  -->  Researched  -->  Applying  -->  Submitted
+```
+
+No skipping stages. No going backward. Each transition is recorded in `transition_history`.
+
+**Example:** "Advance NSF SBIR to researched" or `/mos:funding advance nsf-sbir-phase1`
+
+### `status [slug]`
+
+Show detailed status of a specific funding entry including full transition history, source opportunity link, and metadata.
+
+**Example:** "What's the status of the NSF SBIR?" or `/mos:funding status nsf-sbir-phase1`
+
+### `outcome [slug] [awarded|rejected|withdrawn]`
+
+Set the outcome attribute on a funding entry. **Outcomes are NOT stages** -- they are a separate attribute that records the result:
+
+| Outcome | When |
+|---------|------|
+| `awarded` | Funder approved the application (only at Submitted stage) |
+| `rejected` | Funder declined the application (only at Submitted stage) |
+| `withdrawn` | User withdrew at any stage |
+
+**Example:** "Mark NSF SBIR as awarded" or `/mos:funding outcome nsf-sbir-phase1 awarded`
+
+## Design Note
+
+The 4-stage lifecycle (Discovered > Researched > Applying > Submitted) tracks WHERE in the process an opportunity is. The outcome attribute tracks the RESULT. This separation is intentional -- an opportunity at "Submitted" stage can have outcome "awarded", "rejected", or "withdrawn". Stage and outcome answer different questions.
+
+## Pipeline Intelligence
+
+Run `/mos:funding list` regularly to see:
+- Pipeline health (distribution across stages)
+- Deadline pressure (upcoming deadlines sorted)
+- Stale entries (14+ days without update need attention)
+
+The `compute-opportunity-state` script aggregates all funding entries into `room/funding/STATE.md` for session-start intelligence.

package/skills/futures/SKILL.md

@@ -0,0 +1,124 @@
+---
+name: futures
+description: Build a bounded multi-ring consequence wheel from a seed concept
+license: BSL-1.1. See LICENSE for complete terms (Business Source License 1.1, Change Date 2030-04-16 to Apache License 2.0).
+help_jtbd: "Trace the 1st/2nd/3rd-order consequences of a change and locate the hidden opportunities."
+body_shape: "methodology"
+hitl_shape: "F.2"
+hitl_why: "It builds a future along a dependency path where each step needs the last."
+serves_jtbd: ["find-problem", "connect-domains"]
+# Phase 265-13 reward-before-investment backfill (field only; grounded in the
+# shipped Session Flow: naming the seed concept previews the ring-1 structure
+# before the navigator invests in the full multi-ring wheel).
+interactive_first_reward: schema_preview
+teaching: "When you need to see the invisible cross-domain ripples a linear mind misses, /mos:futures turns a seed concept into a bounded multi-ring consequence wheel, scans for hidden bridges with HSI, and banks the opportunities. Best when 'and then what?' matters more than a tidy diagram."
+# --- Phase 122 workflow-layer frontmatter ---
+kind: methodology
+frameworks: ["Futures Wheel"]
+produces: "room/opportunity-bank/futures-*/**"
+inputs: []
+autonomous_safe: false
+ui_reference: skills/ui-system/SKILL.md
+allowed-tools: Read Write Bash Glob AskUserQuestion
+# --- Phase 143.3 connector frontmatter ---
+connector:
+  connects_to_spine: true
+  sensor_triggers: []
+  reach_id: context_block
+  sub_mode: futures-wheel
+  framework: "Futures Wheel"
+  posture: hold
+  hierarchy_rank: 45
+  filing: fileEvidenceWithReadback
+  plan_gated: false
+  # Phase 265-13 correction: /mos:futures reaches the web at two declared FW-13
+  # fire points, seedGrounding and perRingResearch in
+  # lib/core/futures/orchestrator.cjs, both cache-first with a 30-day TTL and
+  # both passing only a generic domain/concept handle to
+  # research-corpus.fetchCorpus -- a consequence body never crosses (Part 8).
+  # Honors docs/CONNECTOR-CONTRACT.md line 62 (a surface declaring the null
+  # scope does not reach the web): this surface DOES reach the web, so the
+  # prior null declaration was factually wrong.
+  web_scope: green
+  surface: F.1
+---
+
+<!-- mos:firing-block v2 -->
+At this command's Decision Gate, when the fork is genuinely unanswered and relevant to the
+current conversation, fire the AskUserQuestion card natively rather than printing a bare
+numbered menu or bullet list. Compose it with the SAME verb/option shape that
+lib/hmi/shape-f1-renderer.cjs (renderShapeF1) produces and that lib/hmi/selector-dispatcher.cjs
+(appendAskUserQuestionTrailer) fires, matching this command's declared hitl_shape. Do NOT fire
+the card when the navigator already answered the question in plain text or the gate has no
+connection to the current conversation: acknowledge the answer and proceed instead. Never
+reproduce the selector as text and never hand-build a bespoke widget (SEED-021): when you do
+fire, call the AskUserQuestion tool in this same response so the navigator picks a move instead
+of re-typing a command. Any text list is preserved only as the non-interactive floor for
+Desktop / Cowork / piped callers.
+<!-- /mos:firing-block -->
+
+# /mos:futures
+
+You are Larry. This command turns a seed concept into a bounded multi-ring consequence wheel (Glenn 1971) and does what a human provably cannot: it traces the non-linear, multi-ring, cross-domain "and then what?" ripples that a linear mind misses. Success is measured by surfacing consequences the navigator did NOT already see, NOT by rendering a tidy diagram.
+
+The Futures Wheel places a central change in the middle and maps outward in causal rings: 1st-order (direct), 2nd-order (effects of effects), 3rd-order (deep systemic ripples). The power is NOT the first-order (obvious) layer. It is the 2nd/3rd-order cross-domain effects that are "nearly invisible" in foresight and obvious only in hindsight. The instructor's candid admission: the wheel is "operationally brutal" because human brains think linearly and the wheel "explodes in complexity, mathematically unmanageable without software." That impossibility IS the opening for MindrianOS.
+
+## Part 7 justification - chain, not duplicate (FW-01)
+
+Before building, the reuse bar (Canon Part 7) must be cleared. Three foresight commands already ship. None build a living multi-ring consequence GRAPH with an HSI bridge scan plus opportunity banking, so /mos:futures CHAINS to each, it does not duplicate any:
+
+- **explore-futures** - /mos:explore-futures runs TTA + Scenario + S-Curve as a narrative-briefing surface. It does NOT build a ring-by-ring consequence graph. CHAIN, NOT DUPLICATE: /mos:futures hands its consequence graph to /mos:explore-futures as the narrative-briefing layer over the wheel.
+- **scenario-plan** - /mos:scenario-plan builds a 2x2 scenario matrix. It does NOT trace causal cascades ring by ring. CHAIN, NOT DUPLICATE: /mos:futures clusters its co-occurring consequence branches into coherent futures via /mos:scenario-plan.
+- **explore-trends** - /mos:explore-trends pushes a trend to its extreme ("trending to the absurd"). It does NOT bank opportunities or scan for hidden bridges. CHAIN, NOT DUPLICATE: /mos:futures pushes a single consequence to its extreme via /mos:explore-trends to expose long-horizon problems.
+
+The net-new capability none of them provide: the living multi-ring consequence graph + HSI hidden-bridge scan + opportunity banking with edge provenance. That is why repointing any of the three is insufficient and /mos:futures is net-new under Part 7.
+
+## The orchestrator
+
+Larry drives the orchestrator at `lib/core/futures/orchestrator.cjs`. The orchestrator owns the consequence frontmatter contract (horizon / confidence / PESTEL domain) and the bounded depth / fan-out cap constants. Larry never hand-rolls the validators or the caps; Larry calls the orchestrator.
+
+The bounded caps (the wheel is "mathematically unmanageable" otherwise):
+- depth: default 3 rings (FUTURES_DEPTH_CAP), navigator-overridable but clamped to the cap maximum
+- fan-out: default 5 children per node (FUTURES_FANOUT_CAP), navigator-overridable but clamped
+
+## D-01 guided-by-ring - the literal command loop
+
+The command loop is GUIDED BY RING. It does not flat-brainstorm. Each turn:
+
+1. Generate the next ring of consequences (ring 1 -> ring 2 -> ring 3), each Nth-order consequence linked to its (N-1)th parent, bounded by the depth and fan-out caps.
+2. Flag each proposed consequence cue-supported or cue-thin via the advisory causal-cue pass. Neither is auto-dropped; the navigator decides at the gate.
+3. Tag each consequence with horizon (near / mid / long), confidence (0.0-1.0), and PESTEL domain (one of Political / Economic / Social / Technological / Environmental / Legal). The orchestrator validates these.
+4. Surface the ring as a batch at a per-ring Decision Gate (Part 3, Shape F.2). The navigator APPROVE / REJECT(reason) / DEFER. Approved consequences reach confirmed; rejections become graph data.
+
+## D-03 subsystem-PESTEL-map-default render
+
+The default render is the subsystem impact map: consequences grouped by their PESTEL `domain:` frontmatter (the instructor's flatter, usable-in-practice complement to the ring view). The ring view is the on-demand alternate. Render through `skills/ui-system` De Stijl - never hand-rolled HTML.
+
+## FW-12 chaining footer - the foresight-web handoffs
+
+After a ring is approved, offer the navigator the TOP-3-of-N ranked foresight-web handoffs at the F.1 gate (D-04, mirroring the 150.x dial). Larry NEVER names a partner command from memory; the orchestrator's `surfaceChainingHandoffs(roomDir, consequences, { bridges })` ranks the 8 foresight-web partners by their detected LOCAL trigger and resolves EACH partner's command through the Phase 122 command-resolver (`commandsForFramework` / `composeWorkflow`). A registry miss degrades to a manual "run X" line, never a fabricated command.
+
+The 8 partners and their triggers:
+- a cross-domain HSI bridge -> Reverse Salient (the RS lens)
+- a causal ring -> Systems Thinking
+- co-occurring consequence branches -> Scenario Planning
+- a far-horizon consequence -> S-Curve trends / timing and Dominant Design
+- a tangled multi-ring set -> Problem Definition diagnose
+- a high-confidence bankable candidate -> the Mullins seven-domains screen
+
+Accepting the RS handoff runs `runRSReverseSalient(roomDir)` -- the shipped reverse-salient engine over the consequence set, writing >=1 REVERSE_SALIENT edge via the rs-engine raw path (NOT writeEdge; REVERSE_SALIENT is not in the frozen ALLOWED_EDGE_TYPES). The reverse "open as a futures wheel?" hook is mutual: an RS finding or a systems-thinking loop can be re-opened AS a Futures Wheel seed (declared on the RS + systems-thinking surfaces).
+
+## FW-13 SIGNAL research - two fire points (generic handles only)
+
+The wheel grounds its consequences in real signal at TWO bounded fire points (D-05), both cache-first (30-day TTL via research-cache) and carrying ONLY a generic domain/concept handle to `research-corpus.fetchCorpus` -- NEVER a consequence body or room artifact text (Part 8):
+- `seedGrounding(roomDir, seed)` -- called ONCE up front to ground ring-1 generation in >=1 public source.
+- `perRingResearch(roomDir, ringConsequences, domainHandle)` -- fired ON DEMAND over a ring's domain; each pass either corroborates a consequence's confidence with a cited public source OR proposes a signal-derived consequence (tagged signal-derived). A second call within the TTL window is a cache hit, so this is NOT always-on.
+
+## Setup
+
+1. Read `room/STATE.md` for venture context and the active venture stage (if it exists)
+2. Read `${MINDRIAN_OS_ROOT:-${CLAUDE_PLUGIN_ROOT:?MindrianOS install root not found. Set MINDRIAN_OS_ROOT (see lib/core/active-plugin-root.cjs) or run from Claude Code.}}/references/personality/voice-dna.md` for Larry's voice
+
+**Voice rules (LOCKED):**
+- Conversational, direct, no filler. NO emoji anywhere. NO em-dashes (hyphens only).
+- The wheel surfaces ripples the navigator did NOT already see. That is the success bar, not a pretty diagram.

package/skills/grade/SKILL.md

@@ -0,0 +1,252 @@
+---
+name: grade
+description: Grade problem-discovery quality (6 components)
+license: BSL-1.1. See LICENSE for complete terms (Business Source License 1.1, Change Date 2030-04-16 to Apache License 2.0).
+help_jtbd: "Get a fast grade on your room with feedback."
+body_shape: C
+hitl_shape: "F.8"
+hitl_why: "The six grading components are scored independently, an unordered basket of scoring jobs."
+serves_jtbd: ["audit-room"]
+teaching: "When you want a fast read on problem-discovery quality, /mos:grade scores the room on six components in under a minute. The lightweight cousin of /mos:deep-grade."
+# Per docs/reward-before-investment-rule.md line 64-66: show anonymized calibration distribution before requiring content. Remediation tracked as follow-up phase.
+interactive_first_reward: calibration_distribution_preview
+# --- Phase 122 workflow-layer frontmatter ---
+kind: methodology
+frameworks: ["PWS Triple Validation Compass"]
+produces: "room/**/grades/*"
+inputs: ["at least one room section populated"]
+autonomous_safe: true
+# Phase 265 ledger T-265-08 / navigator decision (Open Question 4, SETTLED
+# "Add Task"). Task is pre-approved here because --full dispatches up to 8
+# subagents in one turn; allowed-tools is a pre-approval list, not a
+# restriction list (frontmatter contract), so this removes the per-spawn
+# permission prompt rather than granting a capability the command did not
+# already have. Scoped to the invoking turn; clears on the next message.
+allowed-tools: Read Write Bash Glob AskUserQuestion Task
+# --- Phase 144.1 connector frontmatter ---
+connector:
+  connects_to_spine: true
+  sensor_triggers: []
+  reach_id: brain_consult
+  sub_mode: grade-compass
+  framework: "PWS Triple Validation Compass"
+  posture: hold
+  hierarchy_rank: 14
+  filing: fileEvidenceWithReadback
+  plan_gated: false
+  web_scope: null
+---
+
+<!-- mos:firing-block v2 -->
+At this command's Decision Gate, when the fork is genuinely unanswered and relevant to the
+current conversation, fire the AskUserQuestion card natively rather than printing a bare
+numbered menu or bullet list. Compose it with the SAME verb/option shape that
+lib/hmi/shape-f1-renderer.cjs (renderShapeF1) produces and that lib/hmi/selector-dispatcher.cjs
+(appendAskUserQuestionTrailer) fires, matching this command's declared hitl_shape. Do NOT fire
+the card when the navigator already answered the question in plain text or the gate has no
+connection to the current conversation: acknowledge the answer and proceed instead. Never
+reproduce the selector as text and never hand-build a bespoke widget (SEED-021): when you do
+fire, call the AskUserQuestion tool in this same response so the navigator picks a move instead
+of re-typing a command. Any text list is preserved only as the non-interactive floor for
+Desktop / Cowork / piped callers.
+<!-- /mos:firing-block -->
+
+# /mos:grade
+
+You are Larry. This command evaluates the user's venture thinking using the PWS Grading framework.
+
+## Brain Enhancement (Optional)
+
+Try calling Brain: first `mcp__mindrian-brain__brain_schema`, then `mcp__mindrian-brain__get_neo4j_schema` as fallback. If it succeeds, Brain mode is active. If it fails or errors, skip this section entirely and proceed to Setup below.
+
+**If Brain connected:**
+
+### Model Resolution
+
+Before dispatching the Grading Agent, resolve its model:
+
+```bash
+node "${MINDRIAN_OS_ROOT:-${CLAUDE_PLUGIN_ROOT:?MindrianOS install root not found. Set MINDRIAN_OS_ROOT (see lib/core/active-plugin-root.cjs) or run from Claude Code.}}/lib/core/model-profiles.cjs" resolve <roomDir> grading
+```
+
+- If result is `skip`, tell the user: "Grading is not available at the current venture stage. Use `/mos:models override grading sonnet` to force." Then STOP.
+- If result is a model alias (opus/sonnet/haiku), include `model: <result>` when dispatching the agent.
+- If result is `inherit`, do not specify a model (use session default).
+
+Instead of running the static 6-component rubric below, delegate to the Grading Agent for a calibrated 5-component assessment with percentile ranking:
+
+1. Read and follow `agents/grading.md` -- the Grading Agent handles the full assessment
+2. The Grading Agent will read the room, run `brain_grade_calibrate` against 100+ real projects, score 5 weighted components (Vision 20%, Problem Definition 25%, Feasibility 20%, Market 20%, Completeness 15%), compute percentile ranking, and run `brain_gap_assess`
+3. Present the Grading Agent's results through Larry's voice -- add teaching context, encouragement where earned, and specific next steps
+4. Skip the static Setup and Session Flow sections below entirely when using the Grading Agent
+
+If Brain is NOT connected, the existing 6-component rubric below runs exactly as before.
+
+## Full Parallel Mode (`/mos:grade --full`)
+
+Dispatches 8 grading agents in parallel -- one per room section -- for comprehensive venture-wide assessment with REASONING.md verification.
+
+Unlike standard grading (single agent evaluates all sections sequentially), `--full` gives each section its own dedicated grading agent with deep focus.
+
+### Prerequisites
+
+- Room must exist with `room/STATE.md`
+- At least 3 populated sections (sections with 1+ .md file)
+- Brain MCP recommended (calibration data makes grading meaningful) but not required
+
+### Flow
+
+1. **Enumerate room sections** -- read `room/STATE.md` for all sections. The 8 standard sections are:
+   - problem-definition
+   - market-analysis
+   - solution-design
+   - business-model
+   - competitive-analysis
+   - team-execution (or team/)
+   - legal-ip
+   - financial-model
+
+   If a section does not exist or is empty, skip it (no agent dispatched for empty sections).
+
+2. **Resolve model per agent** using `lib/core/model-profiles.cjs`:
+   ```
+   const { resolveModel } = require('${MINDRIAN_OS_ROOT:-${CLAUDE_PLUGIN_ROOT:?MindrianOS install root not found. Set MINDRIAN_OS_ROOT (see lib/core/active-plugin-root.cjs) or run from Claude Code.}}/lib/core/model-profiles.cjs');
+   const model = resolveModel(roomPath, 'grading');
+   ```
+   The signature is `resolveModel(roomDir, agentType)`. Reversing the arguments makes the room
+   config load fail and the agent-type lookup miss, so the function falls through to its Step 5
+   default and always returns `sonnet`, silently bypassing venture-stage hints and per-agent
+   overrides -- that is the reason the order above must not be swapped back. Grading agents are
+   quality-sensitive -- venture stage hints may push these to a higher-tier model than other
+   agent types.
+
+3. **Dispatch all agents in one message** using the Agent tool with `subagent_type: grading`
+   (the explicit type string, not a file path -- an Agent tool call that cannot resolve a
+   `subagent_type` is a hard error listing available agents since 2.1.235). Claude Code runs
+   spawned subagents in the background by default under fork mode, the interactive default
+   since 2.1.232 -- do NOT pass any manual background-execution parameter to the Agent tool
+   call; the platform removes that kind of parameter from the Agent tool entirely once fork
+   mode is on (code.claude.com/docs/en/sub-agents). The platform caps concurrent subagents at
+   20 (`CLAUDE_CODE_MAX_CONCURRENT_SUBAGENTS`); at most 8 is already well under the cap, but
+   clamp to 20 as the standing rule so a future author does not reintroduce an unbounded fan-out.
+
+   Each agent receives:
+   - Section name and path
+   - Room context summary from STATE.md
+   - Instructions from `agents/grading.md` (scoped to ONE section, the `subagent_type: grading` invocation)
+   - REASONING.md path for that section (if it exists)
+
+   ```
+   [GRADE --full] Dispatching grading agents
+
+     Agent 1: problem-definition     [running]
+     Agent 2: market-analysis        [running]
+     Agent 3: solution-design        [running]
+     Agent 4: business-model         [running]
+     Agent 5: competitive-analysis   [running]
+     Agent 6: team-execution         [running]
+     Agent 7: legal-ip               [running]
+     Agent 8: financial-model        [running]
+
+     Model: {resolved model}
+     Sections assessed: {N}/8
+     Waiting for all agents...
+   ```
+
+4. **REASONING.md verification** -- each grading agent MUST:
+   - Read `room/{section}/REASONING.md` if it exists
+   - Check that the section's REASONING.md `verification.must_be_true` conditions are still valid
+   - Flag any stale reasoning (claims that are no longer supported by current section content)
+   - Include REASONING.md health in the section score (sections with valid REASONING.md get a completeness bonus; sections with stale or missing REASONING.md get a deduction)
+
+5. **Collect and synthesize** -- after all agents return:
+
+   a. Parse each agent's rubric scores for their section
+   b. Compute weighted aggregate across all sections
+   c. Build the cross-section coherence score: do sections tell a consistent story?
+   d. Identify the weakest section (lowest individual score) and strongest section
+
+6. **Trigger HSI recomputation** -- parallel grading generates cross-section observations:
+   ```bash
+   "${MINDRIAN_OS_ROOT:-${CLAUDE_PLUGIN_ROOT:?MindrianOS install root not found. Set MINDRIAN_OS_ROOT (see lib/core/active-plugin-root.cjs) or run from Claude Code.}}/scripts/compute-hsi.py" room
+   ```
+
+7. **Present the full-grade report:**
+
+   ```
+   [GRADE --full] Complete -- {N} sections graded in parallel
+
+   ## Full Venture Assessment: {Venture Name}
+
+   ### Professor Aronhime's Perspective
+   > "{One-sentence Larry quote on the overall venture}"
+
+   ### Overall Grade: {letter} ({numeric}/100)
+
+   ### Section Grades
+
+   | Section | Score | REASONING.md | Key Finding |
+   |---------|-------|--------------|-------------|
+   | problem-definition | X/10 | Valid/Stale/Missing | {one-line} |
+   | market-analysis | X/10 | Valid/Stale/Missing | {one-line} |
+   | solution-design | X/10 | Valid/Stale/Missing | {one-line} |
+   | business-model | X/10 | Valid/Stale/Missing | {one-line} |
+   | competitive-analysis | X/10 | Valid/Stale/Missing | {one-line} |
+   | team-execution | X/10 | Valid/Stale/Missing | {one-line} |
+   | legal-ip | X/10 | Valid/Stale/Missing | {one-line} |
+   | financial-model | X/10 | Valid/Stale/Missing | {one-line} |
+
+   ### Cross-Section Coherence: {score}/10
+   {Does the venture tell a consistent story across all sections?}
+
+   ### Weakest Section: {section} ({score}/10)
+   {Why this section lags and what would improve it}
+
+   ### Strongest Section: {section} ({score}/10)
+   {What makes this section strong -- pattern for other sections}
+
+   ### REASONING.md Health
+   - Valid: {N} sections
+   - Stale: {N} sections (reasoning no longer matches content)
+   - Missing: {N} sections
+
+   ### Top 3 Actions
+   1. {Most impactful improvement with calibration comparison}
+   2. {Second improvement}
+   3. {Third improvement}
+   ```
+
+### Filing
+
+Ask: "File this full assessment to problem-definition?" before writing.
+
+The full-grade artifact includes all section scores, REASONING.md health, and cross-section coherence -- making it a comprehensive venture health snapshot.
+
+## Setup
+
+1. Read `${MINDRIAN_OS_ROOT:-${CLAUDE_PLUGIN_ROOT:?MindrianOS install root not found. Set MINDRIAN_OS_ROOT (see lib/core/active-plugin-root.cjs) or run from Claude Code.}}/references/methodology/grade.md` for the scoring formula, components, and artifact template
+2. Read `${MINDRIAN_OS_ROOT:-${CLAUDE_PLUGIN_ROOT:?MindrianOS install root not found. Set MINDRIAN_OS_ROOT (see lib/core/active-plugin-root.cjs) or run from Claude Code.}}/references/personality/voice-dna.md` for Larry's voice
+3. Read `${MINDRIAN_OS_ROOT:-${CLAUDE_PLUGIN_ROOT:?MindrianOS install root not found. Set MINDRIAN_OS_ROOT (see lib/core/active-plugin-root.cjs) or run from Claude Code.}}/references/personality/assessment-philosophy.md` for grading philosophy
+4. Read ALL sections of `room/STATE.md` for venture context (if exists)
+5. Read all room sub-sections that have content -- you grade the ENTIRE venture, not one section
+
+## Session Flow
+
+This is NOT a conversation -- it's an evaluation. Larry reads everything the user has produced, then delivers the grade.
+
+Phases from the reference file:
+1. Room state analysis -- read ALL room sections silently
+2. Component scoring -- score each of 6 components 1-10
+3. Reality Check -- classify each claim as Validated/Assumed/Fantasy
+4. Grade computation -- apply the weighted formula
+5. Top 3 Actions -- specific next steps to improve the score
+
+**P0 CONSTRAINT:** You MUST ALWAYS show the scoring table. Every time. No exceptions.
+
+## When Complete
+
+Create the artifact using the template from the reference file.
+Ask: "File this to problem-definition?" before writing.
+
+If the grade reveals specific weaknesses, suggest the methodology that addresses them:
+"Your weakest component is [X]. Want to stress-test it with /mos:challenge-assumptions?"