[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7CCT3P89K00T175RFMQW0WN":3,"skill-same:01M7CCT3P89K00T175RFMQW0WN":57},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":31,"commitSha":32,"skillDirectory":33,"forksCount":14,"repositoryHtmlUrl":34,"repositoryHomepageUrl":6,"ownerHtmlUrl":35,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":36,"install":37,"resources":50},"01M7CCT3P89K00T175RFMQW0WN","agent-protocol",null,"Inter-agent communication protocol for C-suite agent teams. Defines invocation syntax, loop prevention, isolation rules, and response formats. Use when C-suite agents need to query each other, coordinate cross-functional analysis, or run board meetings with multiple agent roles.","azhri1990\u002Fjarvis-mega-repo","azhri1990","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F281207543?v=4","01M7BMC0WHXB2ZA9SH1PFJDYNS","Nashriq Mokhtar","01M7C4WMQVQ450E0NA99BG7BHA",0,"MIT","https:\u002F\u002Fgithub.com\u002Fazhri1990\u002Fjarvis-mega-repo\u002Fblob\u002Fmain\u002Frepos\u002FJ.A.R.V.I.S\u002Fcomponents\u002Fcore\u002Fclaude-skills\u002F.vibe\u002Fskills\u002Fclaude-skills\u002Fc-level-advisor\u002Fagent-protocol\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: \"agent-protocol\"\ndescription: \"Inter-agent communication protocol for C-suite agent teams. Defines invocation syntax, loop prevention, isolation rules, and response formats. Use when C-suite agents need to query each other, coordinate cross-functional analysis, or run board meetings with multiple agent roles.\"\nlicense: MIT\nmetadata:\n  version: 1.0.0\n  author: Alireza Rezvani\n  category: c-level\n  domain: agent-orchestration\n  updated: 2026-03-05\n  frameworks: invocation-patterns\n---\n\n# Inter-Agent Protocol\n\nHow C-suite agents talk to each other. Rules that prevent chaos, loops, and circular reasoning.\n\n## Keywords\nagent protocol, inter-agent communication, agent invocation, agent orchestration, multi-agent, c-suite coordination, agent chain, loop prevention, agent isolation, board meeting protocol\n\n## Invocation Syntax\n\nAny agent can query another using:\n\n```\n[INVOKE:role|question]\n```\n\n**Examples:**\n```\n[INVOKE:cfo|What's the burn rate impact of hiring 5 engineers in Q3?]\n[INVOKE:cto|Can we realistically ship this feature by end of quarter?]\n[INVOKE:chro|What's our typical time-to-hire for senior engineers?]\n[INVOKE:cro|What does our pipeline look like for the next 90 days?]\n```\n\n**Valid roles:** `ceo`, `cfo`, `cro`, `cmo`, `cpo`, `cto`, `chro`, `coo`, `ciso`, `gc`, `cdo`, `caio`, `cco`, `vpe`\n\n| Role token | Advisor skill |\n|---|---|\n| `gc` | general-counsel-advisor (legal, contracts, term sheets) |\n| `cdo` | chief-data-officer-advisor (data strategy, training-data rights) |\n| `caio` | chief-ai-officer-advisor (AI strategy, evals, AI risk) |\n| `cco` | chief-customer-officer-advisor (retention, customer success) |\n| `vpe` | vpe-advisor (engineering delivery, DORA, eng hiring) |\n\n## Response Format\n\nInvoked agents respond using this structure:\n\n```\n[RESPONSE:role]\nKey finding: [one line — the actual answer]\nSupporting data:\n  - [data point 1]\n  - [data point 2]\n  - [data point 3 — optional]\nConfidence: [high | medium | low]\nCaveat: [one line — what could make this wrong]\n[\u002FRESPONSE]\n```\n\n**Example:**\n```\n[RESPONSE:cfo]\nKey finding: Hiring 5 engineers in Q3 extends runway from 14 to 9 months at current burn.\nSupporting data:\n  - Current monthly burn: $280K → increases to ~$380K (+$100K fully loaded)\n  - ARR needed to offset: ~$1.2M additional within 12 months\n  - Current pipeline covers 60% of that target\nConfidence: medium\nCaveat: Assumes 3-month ramp and no change in revenue trajectory.\n[\u002FRESPONSE]\n```\n\n## Loop Prevention (Hard Rules)\n\nThese rules are enforced unconditionally. No exceptions.\n\n### Rule 1: No Self-Invocation\nAn agent cannot invoke itself.\n```\n❌ CFO → [INVOKE:cfo|...] — BLOCKED\n```\n\n### Rule 2: Maximum Depth = 2\nChains can go A→B→C. The third hop is blocked.\n```\n✅ CRO → CFO → COO (depth 2)\n❌ CRO → CFO → COO → CHRO (depth 3 — BLOCKED)\n```\n\n### Rule 3: No Circular Calls\nIf agent A called agent B, agent B cannot call agent A in the same chain.\n```\n✅ CRO → CFO → CMO\n❌ CRO → CFO → CRO (circular — BLOCKED)\n```\n\n### Rule 4: Chain Tracking\nEach invocation carries its call chain. Format:\n```\n[CHAIN: cro → cfo → coo]\n```\nAgents check this chain before responding with another invocation.\n\n**When blocked:** Return this instead of invoking:\n```\n[BLOCKED: cannot invoke cfo — circular call detected in chain cro→cfo]\nState assumption used instead: [explicit assumption the agent is making]\n```\n\n## Isolation Rules\n\n### Board Meeting Phase 2 (Independent Analysis)\n**NO invocations allowed.** Each role forms independent views before cross-pollination.\n- Reason: prevent anchoring and groupthink\n- Duration: entire Phase 2 analysis period\n- If an agent needs data from another role: state explicit assumption, flag it with `[ASSUMPTION: ...]`\n\n### Board Meeting Phase 3 (Critic Role)\nExecutive Mentor can **reference** other roles' outputs but **cannot invoke** them.\n- Reason: critique must be independent of new data requests\n- Allowed: \"The CFO's projection assumes X, which contradicts the CRO's pipeline data\"\n- Not allowed: `[INVOKE:cfo|...]` during critique phase\n\n### Outside Board Meetings\nInvocations are allowed freely, subject to loop prevention rules above.\n\n## When to Invoke vs When to Assume\n\n**Invoke when:**\n- The question requires domain-specific data you don't have\n- An error here would materially change the recommendation\n- The question is cross-functional by nature (e.g., hiring impact on both budget and capacity)\n\n**Assume when:**\n- The data is directionally clear and precision isn't critical\n- You're in Phase 2 isolation (always assume, never invoke)\n- The chain is already at depth 2\n- The question is minor compared to your main analysis\n\n**When assuming, always state it:**\n```\n[ASSUMPTION: runway ~12 months based on typical Series A burn profile — not verified with CFO]\n```\n\n## Conflict Resolution\n\nWhen two invoked agents give conflicting answers:\n\n1. **Flag the conflict explicitly:**\n   ```\n   [CONFLICT: CFO projects 14-month runway; CRO expects pipeline to close 80% → implies 18+ months]\n   ```\n2. **State the resolution approach:**\n   - Conservative: use the worse case\n   - Probabilistic: weight by confidence scores\n   - Escalate: flag for human decision\n3. **Never silently pick one** — surface the conflict to the user.\n\n## Broadcast Pattern (Crisis \u002F CEO)\n\nCEO can broadcast to all roles simultaneously:\n```\n[BROADCAST:all|What's the impact if we miss the fundraise?]\n```\n\nResponses come back independently (no agent sees another's response before forming its own). Aggregate after all respond.\n\n## Decision Memory (Canonical Layout)\n\nAll C-suite skills and `\u002Fcs:*` commands read and write decisions in **one** place — the two-layer model owned by `\u002Fcs:decide` and the decision-logger skill:\n\n```\n~\u002F.claude\u002Fdecisions\u002F\n├── raw\u002FYYYY-MM-DD-\u003Cslug>.md        # Layer 1 — full transcripts\u002Fdeliberations (never auto-loaded)\n├── raw\u002Farchive\u002FYYYY\u002F               # Raw files after 90 days\n├── approved\u002FYYYY-MM-DD-\u003Cslug>.md   # Layer 2 — one founder-approved decision record per file\n└── approved\u002Fdecisions.md           # Layer 2 index — append-only log of approved decisions\n```\n\n**Rules:**\n- **Layer 1 (raw)** stores everything, including rejected arguments. Reference only — never feeds future sessions automatically.\n- **Layer 2 (approved)** stores only founder-approved decisions. This is what board meetings, `\u002Fcs:office-hours`, and `\u002Fcs:founder-mode` load. Prevents hallucinated consensus.\n- Writers: `\u002Fcs:decide` and the Chief of Staff (post board-meeting Phase 5). Individual role agents never write decisions directly.\n- decision-logger, chief-of-staff, and board-meeting all use this layout. Their SKILL.md files link here rather than defining their own paths.\n\n**Migration:** earlier versions used `memory\u002Fboard-meetings\u002F` (decision-logger, board-meeting) and `~\u002F.claude\u002Fdecision-log.md` (chief-of-staff); read those for history if present, but write all new entries to `~\u002F.claude\u002Fdecisions\u002F`.\n\n## Quick Reference\n\n| Rule | Behavior |\n|------|----------|\n| Self-invoke | ❌ Always blocked |\n| Depth > 2 | ❌ Blocked, state assumption |\n| Circular | ❌ Blocked, state assumption |\n| Phase 2 isolation | ❌ No invocations |\n| Phase 3 critique | ❌ Reference only, no invoke |\n| Conflict | ✅ Surface it, don't hide it |\n| Assumption | ✅ Always explicit with `[ASSUMPTION: ...]` |\n\n## Internal Quality Loop (before anything reaches the founder)\n\nNo role presents to the founder without passing through this verification loop. The founder sees polished, verified output — not first drafts.\n\n### Step 1: Self-Verification (every role, every time)\n\nBefore presenting, every role runs this internal checklist:\n\n```\nSELF-VERIFY CHECKLIST:\n□ Source Attribution — Where did each data point come from?\n  ✅ \"ARR is $2.1M (from CRO pipeline report, Q4 actuals)\"\n  ❌ \"ARR is around $2M\" (no source, vague)\n\n□ Assumption Audit — What am I assuming vs what I verified?\n  Tag every assumption: [VERIFIED: checked against data] or [ASSUMED: not verified]\n  If >50% of findings are ASSUMED → flag low confidence\n\n□ Confidence Score — How sure am I on each finding?\n  🟢 High: verified data, established pattern, multiple sources\n  🟡 Medium: single source, reasonable inference, some uncertainty\n  🔴 Low: assumption-based, limited data, first-time analysis\n\n□ Contradiction Check — Does this conflict with known context?\n  Check against company-context.md and recent decisions in decision-log\n  If it contradicts a past decision → flag explicitly\n\n□ \"So What?\" Test — Does every finding have a business consequence?\n  If you can't answer \"so what?\" in one sentence → cut it\n```\n\n### Step 2: Peer Verification (cross-functional validation)\n\nWhen a recommendation impacts another role's domain, that role validates BEFORE presenting.\n\n| If your recommendation involves... | Validate with... | They check... |\n|-------------------------------------|-------------------|---------------|\n| Financial numbers or budget | CFO | Math, runway impact, budget reality |\n| Revenue projections | CRO | Pipeline backing, historical accuracy |\n| Headcount or hiring | CHRO | Market reality, comp feasibility, timeline |\n| Technical feasibility or timeline | CTO | Engineering capacity, technical debt load |\n| Operational process changes | COO | Capacity, dependencies, scaling impact |\n| Customer-facing changes | CRO + CPO | Churn risk, product roadmap conflict |\n| Security or compliance claims | CISO | Actual posture, regulation requirements |\n| Market or positioning claims | CMO | Data backing, competitive reality |\n| Legal exposure, contracts, term sheets | GC | Clause risk, IP ownership, regulatory triggers |\n| Data rights, training-data provenance | CDO | Consent basis, GDPR Art. 6, data-asset impact |\n| AI model claims, eval results, AI risk | CAIO | Eval coverage, hallucination SLO, EU AI Act tier |\n| Retention, churn, customer-health claims | CCO | GRR\u002FNRR decomposition, churn root cause |\n| Delivery timelines, eng throughput | VPE | DORA metrics, cycle-time reality, team capacity |\n\n**Peer validation format:**\n```\n[PEER-VERIFY:cfo]\nValidated: ✅ Burn rate calculation correct\nAdjusted: ⚠️ Hiring timeline should be Q3 not Q2 (budget constraint)\nFlagged: 🔴 Missing equity cost in total comp projection\n[\u002FPEER-VERIFY]\n```\n\n**Skip peer verification when:**\n- Single-domain question with no cross-functional impact\n- Time-sensitive proactive alert (send alert, verify after)\n- Founder explicitly asked for a quick take\n\n### Step 3: Critic Pre-Screen (high-stakes decisions only)\n\nFor decisions that are **irreversible, high-cost, or bet-the-company**, the Executive Mentor pre-screens before the founder sees it.\n\n**Triggers for pre-screen:**\n- Involves spending > 20% of remaining runway\n- Affects >30% of the team (layoffs, reorg)\n- Changes company strategy or direction\n- Involves external commitments (fundraising terms, partnerships, M&A)\n- Any recommendation where all roles agree (suspicious consensus)\n\n**Pre-screen output:**\n```\n[CRITIC-SCREEN]\nWeakest point: [The single biggest vulnerability in this recommendation]\nMissing perspective: [What nobody considered]\nIf wrong, the cost is: [Quantified downside]\nProceed: ✅ With noted risks | ⚠️ After addressing [specific gap] | 🔴 Rethink\n[\u002FCRITIC-SCREEN]\n```\n\n### Step 4: Course Correction (after founder feedback)\n\nThe loop doesn't end at delivery. After the founder responds:\n\n```\nFOUNDER FEEDBACK LOOP:\n1. Founder approves → log decision (Layer 2), assign actions\n2. Founder modifies → update analysis with corrections, re-verify changed parts\n3. Founder rejects → log rejection with DO_NOT_RESURFACE, understand WHY\n4. Founder asks follow-up → deepen analysis on specific point, re-verify\n\nPOST-DECISION REVIEW (30\u002F60\u002F90 days):\n- Was the recommendation correct?\n- What did we miss?\n- Update company-context.md with what we learned\n- If wrong → document the lesson, adjust future analysis\n```\n\n### Verification Level by Stakes\n\n| Stakes | Self-Verify | Peer-Verify | Critic Pre-Screen |\n|--------|-------------|-------------|-------------------|\n| Low (informational) | ✅ Required | ❌ Skip | ❌ Skip |\n| Medium (operational) | ✅ Required | ✅ Required | ❌ Skip |\n| High (strategic) | ✅ Required | ✅ Required | ✅ Required |\n| Critical (irreversible) | ✅ Required | ✅ Required | ✅ Required + board meeting |\n\n### What Changes in the Output Format\n\nThe verified output adds confidence and source information:\n\n```\nBOTTOM LINE\n[Answer] — Confidence: 🟢 High\n\nWHAT\n• [Finding 1] [VERIFIED: Q4 actuals] 🟢\n• [Finding 2] [VERIFIED: CRO pipeline data] 🟢  \n• [Finding 3] [ASSUMED: based on industry benchmarks] 🟡\n\nPEER-VERIFIED BY: CFO (math ✅), CTO (timeline ⚠️ adjusted to Q3)\n```\n\n---\n\n## User Communication Standard\n\nAll C-suite output to the founder follows ONE format. No exceptions. The founder is the decision-maker — give them results, not process.\n\n### Standard Output (single-role response)\n\n```\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\n📊 [ROLE] — [Topic]\n\nBOTTOM LINE\n[One sentence. The answer. No preamble.]\n\nWHAT\n• [Finding 1 — most critical]\n• [Finding 2]\n• [Finding 3]\n(Max 5 bullets. If more needed → reference doc.)\n\nWHY THIS MATTERS\n[1-2 sentences. Business impact. Not theory — consequence.]\n\nHOW TO ACT\n1. [Action] → [Owner] → [Deadline]\n2. [Action] → [Owner] → [Deadline]\n3. [Action] → [Owner] → [Deadline]\n\n⚠️ RISKS (if any)\n• [Risk + what triggers it]\n\n🔑 YOUR DECISION (if needed)\nOption A: [Description] — [Trade-off]\nOption B: [Description] — [Trade-off]\nRecommendation: [Which and why, in one line]\n\n📎 DETAIL: [reference doc or script output for deep-dive]\n\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n```\n\n### Proactive Alert (unsolicited — triggered by context)\n\n```\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\n🚩 [ROLE] — Proactive Alert\n\nWHAT I NOTICED\n[What triggered this — specific, not vague]\n\nWHY IT MATTERS\n[Business consequence if ignored — in dollars, time, or risk]\n\nRECOMMENDED ACTION\n[Exactly what to do, who does it, by when]\n\nURGENCY: 🔴 Act today | 🟡 This week | ⚪ Next review\n\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n```\n\n### Board Meeting Output (multi-role synthesis)\n\n```\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\n📋 BOARD MEETING — [Date] — [Agenda Topic]\n\nDECISION REQUIRED\n[Frame the decision in one sentence]\n\nPERSPECTIVES\n  CEO: [one-line position]\n  CFO: [one-line position]\n  CRO: [one-line position]\n  [... only roles that contributed]\n\nWHERE THEY AGREE\n• [Consensus point 1]\n• [Consensus point 2]\n\nWHERE THEY DISAGREE\n• [Conflict] — CEO says X, CFO says Y\n• [Conflict] — CRO says X, CPO says Y\n\nCRITIC'S VIEW (Executive Mentor)\n[The uncomfortable truth nobody else said]\n\nRECOMMENDED DECISION\n[Clear recommendation with rationale]\n\nACTION ITEMS\n1. [Action] → [Owner] → [Deadline]\n2. [Action] → [Owner] → [Deadline]\n3. [Action] → [Owner] → [Deadline]\n\n🔑 YOUR CALL\n[Options if you disagree with the recommendation]\n\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n```\n\n### Communication Rules (non-negotiable)\n\n1. **Bottom line first.** Always. The founder's time is the scarcest resource.\n2. **Results and decisions only.** No process narration (\"First I analyzed...\"). No thinking out loud.\n3. **What + Why + How.** Every finding explains WHAT it is, WHY it matters (business impact), and HOW to act on it.\n4. **Max 5 bullets per section.** Longer = reference doc.\n5. **Actions have owners and deadlines.** \"We should consider\" is banned. Who does what by when.\n6. **Decisions framed as options.** Not \"what do you think?\" — \"Option A or B, here's the trade-off, here's my recommendation.\"\n7. **The founder decides.** Roles recommend. The founder approves, modifies, or rejects. Every output respects this hierarchy.\n8. **Risks are concrete.** Not \"there might be risks\" — \"if X happens, Y breaks, costing $Z.\"\n9. **No jargon without explanation.** If you use a term, explain it on first use.\n10. **Silence is an option.** If there's nothing to report, don't fabricate updates.\n\n## Reference\n- `references\u002Finvocation-patterns.md` — common cross-functional patterns with examples\n","# Inter-Agent Protocol\n\nHow C-suite agents talk to each other. Rules that prevent chaos, loops, and circular reasoning.\n\n## Keywords\nagent protocol, inter-agent communication, agent invocation, agent orchestration, multi-agent, c-suite coordination, agent chain, loop prevention, agent isolation, board meeting protocol\n\n## Invocation Syntax\n\nAny agent can query another using:\n\n```\n[INVOKE:role|question]\n```\n\n**Examples:**\n```\n[INVOKE:cfo|What's the burn rate impact of hiring 5 engineers in Q3?]\n[INVOKE:cto|Can we realistically ship this feature by end of quarter?]\n[INVOKE:chro|What's our typical time-to-hire for senior engineers?]\n[INVOKE:cro|What does our pipeline look like for the next 90 days?]\n```\n\n**Valid roles:** `ceo`, `cfo`, `cro`, `cmo`, `cpo`, `cto`, `chro`, `coo`, `ciso`, `gc`, `cdo`, `caio`, `cco`, `vpe`\n\n| Role token | Advisor skill |\n|---|---|\n| `gc` | general-counsel-advisor (legal, contracts, term sheets) |\n| `cdo` | chief-data-officer-advisor (data strategy, training-data rights) |\n| `caio` | chief-ai-officer-advisor (AI strategy, evals, AI risk) |\n| `cco` | chief-customer-officer-advisor (retention, customer success) |\n| `vpe` | vpe-advisor (engineering delivery, DORA, eng hiring) |\n\n## Response Format\n\nInvoked agents respond using this structure:\n\n```\n[RESPONSE:role]\nKey finding: [one line — the actual answer]\nSupporting data:\n  - [data point 1]\n  - [data point 2]\n  - [data point 3 — optional]\nConfidence: [high | medium | low]\nCaveat: [one line — what could make this wrong]\n[\u002FRESPONSE]\n```\n\n**Example:**\n```\n[RESPONSE:cfo]\nKey finding: Hiring 5 engineers in Q3 extends runway from 14 to 9 months at current burn.\nSupporting data:\n  - Current monthly burn: $280K → increases to ~$380K (+$100K fully loaded)\n  - ARR needed to offset: ~$1.2M additional within 12 months\n  - Current pipeline covers 60% of that target\nConfidence: medium\nCaveat: Assumes 3-month ramp and no change in revenue trajectory.\n[\u002FRESPONSE]\n```\n\n## Loop Prevention (Hard Rules)\n\nThese rules are enforced unconditionally. No exceptions.\n\n### Rule 1: No Self-Invocation\nAn agent cannot invoke itself.\n```\n❌ CFO → [INVOKE:cfo|...] — BLOCKED\n```\n\n### Rule 2: Maximum Depth = 2\nChains can go A→B→C. The third hop is blocked.\n```\n✅ CRO → CFO → COO (depth 2)\n❌ CRO → CFO → COO → CHRO (depth 3 — BLOCKED)\n```\n\n### Rule 3: No Circular Calls\nIf agent A called agent B, agent B cannot call agent A in the same chain.\n```\n✅ CRO → CFO → CMO\n❌ CRO → CFO → CRO (circular — BLOCKED)\n```\n\n### Rule 4: Chain Tracking\nEach invocation carries its call chain. Format:\n```\n[CHAIN: cro → cfo → coo]\n```\nAgents check this chain before responding with another invocation.\n\n**When blocked:** Return this instead of invoking:\n```\n[BLOCKED: cannot invoke cfo — circular call detected in chain cro→cfo]\nState assumption used instead: [explicit assumption the agent is making]\n```\n\n## Isolation Rules\n\n### Board Meeting Phase 2 (Independent Analysis)\n**NO invocations allowed.** Each role forms independent views before cross-pollination.\n- Reason: prevent anchoring and groupthink\n- Duration: entire Phase 2 analysis period\n- If an agent needs data from another role: state explicit assumption, flag it with `[ASSUMPTION: ...]`\n\n### Board Meeting Phase 3 (Critic Role)\nExecutive Mentor can **reference** other roles' outputs but **cannot invoke** them.\n- Reason: critique must be independent of new data requests\n- Allowed: \"The CFO's projection assumes X, which contradicts the CRO's pipeline data\"\n- Not allowed: `[INVOKE:cfo|...]` during critique phase\n\n### Outside Board Meetings\nInvocations are allowed freely, subject to loop prevention rules above.\n\n## When to Invoke vs When to Assume\n\n**Invoke when:**\n- The question requires domain-specific data you don't have\n- An error here would materially change the recommendation\n- The question is cross-functional by nature (e.g., hiring impact on both budget and capacity)\n\n**Assume when:**\n- The data is directionally clear and precision isn't critical\n- You're in Phase 2 isolation (always assume, never invoke)\n- The chain is already at depth 2\n- The question is minor compared to your main analysis\n\n**When assuming, always state it:**\n```\n[ASSUMPTION: runway ~12 months based on typical Series A burn profile — not verified with CFO]\n```\n\n## Conflict Resolution\n\nWhen two invoked agents give conflicting answers:\n\n1. **Flag the conflict explicitly:**\n   ```\n   [CONFLICT: CFO projects 14-month runway; CRO expects pipeline to close 80% → implies 18+ months]\n   ```\n2. **State the resolution approach:**\n   - Conservative: use the worse case\n   - Probabilistic: weight by confidence scores\n   - Escalate: flag for human decision\n3. **Never silently pick one** — surface the conflict to the user.\n\n## Broadcast Pattern (Crisis \u002F CEO)\n\nCEO can broadcast to all roles simultaneously:\n```\n[BROADCAST:all|What's the impact if we miss the fundraise?]\n```\n\nResponses come back independently (no agent sees another's response before forming its own). Aggregate after all respond.\n\n## Decision Memory (Canonical Layout)\n\nAll C-suite skills and `\u002Fcs:*` commands read and write decisions in **one** place — the two-layer model owned by `\u002Fcs:decide` and the decision-logger skill:\n\n```\n~\u002F.claude\u002Fdecisions\u002F\n├── raw\u002FYYYY-MM-DD-\u003Cslug>.md        # Layer 1 — full transcripts\u002Fdeliberations (never auto-loaded)\n├── raw\u002Farchive\u002FYYYY\u002F               # Raw files after 90 days\n├── approved\u002FYYYY-MM-DD-\u003Cslug>.md   # Layer 2 — one founder-approved decision record per file\n└── approved\u002Fdecisions.md           # Layer 2 index — append-only log of approved decisions\n```\n\n**Rules:**\n- **Layer 1 (raw)** stores everything, including rejected arguments. Reference only — never feeds future sessions automatically.\n- **Layer 2 (approved)** stores only founder-approved decisions. This is what board meetings, `\u002Fcs:office-hours`, and `\u002Fcs:founder-mode` load. Prevents hallucinated consensus.\n- Writers: `\u002Fcs:decide` and the Chief of Staff (post board-meeting Phase 5). Individual role agents never write decisions directly.\n- decision-logger, chief-of-staff, and board-meeting all use this layout. Their SKILL.md files link here rather than defining their own paths.\n\n**Migration:** earlier versions used `memory\u002Fboard-meetings\u002F` (decision-logger, board-meeting) and `~\u002F.claude\u002Fdecision-log.md` (chief-of-staff); read those for history if present, but write all new entries to `~\u002F.claude\u002Fdecisions\u002F`.\n\n## Quick Reference\n\n| Rule | Behavior |\n|------|----------|\n| Self-invoke | ❌ Always blocked |\n| Depth > 2 | ❌ Blocked, state assumption |\n| Circular | ❌ Blocked, state assumption |\n| Phase 2 isolation | ❌ No invocations |\n| Phase 3 critique | ❌ Reference only, no invoke |\n| Conflict | ✅ Surface it, don't hide it |\n| Assumption | ✅ Always explicit with `[ASSUMPTION: ...]` |\n\n## Internal Quality Loop (before anything reaches the founder)\n\nNo role presents to the founder without passing through this verification loop. The founder sees polished, verified output — not first drafts.\n\n### Step 1: Self-Verification (every role, every time)\n\nBefore presenting, every role runs this internal checklist:\n\n```\nSELF-VERIFY CHECKLIST:\n□ Source Attribution — Where did each data point come from?\n  ✅ \"ARR is $2.1M (from CRO pipeline report, Q4 actuals)\"\n  ❌ \"ARR is around $2M\" (no source, vague)\n\n□ Assumption Audit — What am I assuming vs what I verified?\n  Tag every assumption: [VERIFIED: checked against data] or [ASSUMED: not verified]\n  If >50% of findings are ASSUMED → flag low confidence\n\n□ Confidence Score — How sure am I on each finding?\n  🟢 High: verified data, established pattern, multiple sources\n  🟡 Medium: single source, reasonable inference, some uncertainty\n  🔴 Low: assumption-based, limited data, first-time analysis\n\n□ Contradiction Check — Does this conflict with known context?\n  Check against company-context.md and recent decisions in decision-log\n  If it contradicts a past decision → flag explicitly\n\n□ \"So What?\" Test — Does every finding have a business consequence?\n  If you can't answer \"so what?\" in one sentence → cut it\n```\n\n### Step 2: Peer Verification (cross-functional validation)\n\nWhen a recommendation impacts another role's domain, that role validates BEFORE presenting.\n\n| If your recommendation involves... | Validate with... | They check... |\n|-------------------------------------|-------------------|---------------|\n| Financial numbers or budget | CFO | Math, runway impact, budget reality |\n| Revenue projections | CRO | Pipeline backing, historical accuracy |\n| Headcount or hiring | CHRO | Market reality, comp feasibility, timeline |\n| Technical feasibility or timeline | CTO | Engineering capacity, technical debt load |\n| Operational process changes | COO | Capacity, dependencies, scaling impact |\n| Customer-facing changes | CRO + CPO | Churn risk, product roadmap conflict |\n| Security or compliance claims | CISO | Actual posture, regulation requirements |\n| Market or positioning claims | CMO | Data backing, competitive reality |\n| Legal exposure, contracts, term sheets | GC | Clause risk, IP ownership, regulatory triggers |\n| Data rights, training-data provenance | CDO | Consent basis, GDPR Art. 6, data-asset impact |\n| AI model claims, eval results, AI risk | CAIO | Eval coverage, hallucination SLO, EU AI Act tier |\n| Retention, churn, customer-health claims | CCO | GRR\u002FNRR decomposition, churn root cause |\n| Delivery timelines, eng throughput | VPE | DORA metrics, cycle-time reality, team capacity |\n\n**Peer validation format:**\n```\n[PEER-VERIFY:cfo]\nValidated: ✅ Burn rate calculation correct\nAdjusted: ⚠️ Hiring timeline should be Q3 not Q2 (budget constraint)\nFlagged: 🔴 Missing equity cost in total comp projection\n[\u002FPEER-VERIFY]\n```\n\n**Skip peer verification when:**\n- Single-domain question with no cross-functional impact\n- Time-sensitive proactive alert (send alert, verify after)\n- Founder explicitly asked for a quick take\n\n### Step 3: Critic Pre-Screen (high-stakes decisions only)\n\nFor decisions that are **irreversible, high-cost, or bet-the-company**, the Executive Mentor pre-screens before the founder sees it.\n\n**Triggers for pre-screen:**\n- Involves spending > 20% of remaining runway\n- Affects >30% of the team (layoffs, reorg)\n- Changes company strategy or direction\n- Involves external commitments (fundraising terms, partnerships, M&A)\n- Any recommendation where all roles agree (suspicious consensus)\n\n**Pre-screen output:**\n```\n[CRITIC-SCREEN]\nWeakest point: [The single biggest vulnerability in this recommendation]\nMissing perspective: [What nobody considered]\nIf wrong, the cost is: [Quantified downside]\nProceed: ✅ With noted risks | ⚠️ After addressing [specific gap] | 🔴 Rethink\n[\u002FCRITIC-SCREEN]\n```\n\n### Step 4: Course Correction (after founder feedback)\n\nThe loop doesn't end at delivery. After the founder responds:\n\n```\nFOUNDER FEEDBACK LOOP:\n1. Founder approves → log decision (Layer 2), assign actions\n2. Founder modifies → update analysis with corrections, re-verify changed parts\n3. Founder rejects → log rejection with DO_NOT_RESURFACE, understand WHY\n4. Founder asks follow-up → deepen analysis on specific point, re-verify\n\nPOST-DECISION REVIEW (30\u002F60\u002F90 days):\n- Was the recommendation correct?\n- What did we miss?\n- Update company-context.md with what we learned\n- If wrong → document the lesson, adjust future analysis\n```\n\n### Verification Level by Stakes\n\n| Stakes | Self-Verify | Peer-Verify | Critic Pre-Screen |\n|--------|-------------|-------------|-------------------|\n| Low (informational) | ✅ Required | ❌ Skip | ❌ Skip |\n| Medium (operational) | ✅ Required | ✅ Required | ❌ Skip |\n| High (strategic) | ✅ Required | ✅ Required | ✅ Required |\n| Critical (irreversible) | ✅ Required | ✅ Required | ✅ Required + board meeting |\n\n### What Changes in the Output Format\n\nThe verified output adds confidence and source information:\n\n```\nBOTTOM LINE\n[Answer] — Confidence: 🟢 High\n\nWHAT\n• [Finding 1] [VERIFIED: Q4 actuals] 🟢\n• [Finding 2] [VERIFIED: CRO pipeline data] 🟢  \n• [Finding 3] [ASSUMED: based on industry benchmarks] 🟡\n\nPEER-VERIFIED BY: CFO (math ✅), CTO (timeline ⚠️ adjusted to Q3)\n```\n\n---\n\n## User Communication Standard\n\nAll C-suite output to the founder follows ONE format. No exceptions. The founder is the decision-maker — give them results, not process.\n\n### Standard Output (single-role response)\n\n```\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\n📊 [ROLE] — [Topic]\n\nBOTTOM LINE\n[One sentence. The answer. No preamble.]\n\nWHAT\n• [Finding 1 — most critical]\n• [Finding 2]\n• [Finding 3]\n(Max 5 bullets. If more needed → reference doc.)\n\nWHY THIS MATTERS\n[1-2 sentences. Business impact. Not theory — consequence.]\n\nHOW TO ACT\n1. [Action] → [Owner] → [Deadline]\n2. [Action] → [Owner] → [Deadline]\n3. [Action] → [Owner] → [Deadline]\n\n⚠️ RISKS (if any)\n• [Risk + what triggers it]\n\n🔑 YOUR DECISION (if needed)\nOption A: [Description] — [Trade-off]\nOption B: [Description] — [Trade-off]\nRecommendation: [Which and why, in one line]\n\n📎 DETAIL: [reference doc or script output for deep-dive]\n\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n```\n\n### Proactive Alert (unsolicited — triggered by context)\n\n```\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\n🚩 [ROLE] — Proactive Alert\n\nWHAT I NOTICED\n[What triggered this — specific, not vague]\n\nWHY IT MATTERS\n[Business consequence if ignored — in dollars, time, or risk]\n\nRECOMMENDED ACTION\n[Exactly what to do, who does it, by when]\n\nURGENCY: 🔴 Act today | 🟡 This week | ⚪ Next review\n\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n```\n\n### Board Meeting Output (multi-role synthesis)\n\n```\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n\n📋 BOARD MEETING — [Date] — [Agenda Topic]\n\nDECISION REQUIRED\n[Frame the decision in one sentence]\n\nPERSPECTIVES\n  CEO: [one-line position]\n  CFO: [one-line position]\n  CRO: [one-line position]\n  [... only roles that contributed]\n\nWHERE THEY AGREE\n• [Consensus point 1]\n• [Consensus point 2]\n\nWHERE THEY DISAGREE\n• [Conflict] — CEO says X, CFO says Y\n• [Conflict] — CRO says X, CPO says Y\n\nCRITIC'S VIEW (Executive Mentor)\n[The uncomfortable truth nobody else said]\n\nRECOMMENDED DECISION\n[Clear recommendation with rationale]\n\nACTION ITEMS\n1. [Action] → [Owner] → [Deadline]\n2. [Action] → [Owner] → [Deadline]\n3. [Action] → [Owner] → [Deadline]\n\n🔑 YOUR CALL\n[Options if you disagree with the recommendation]\n\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n```\n\n### Communication Rules (non-negotiable)\n\n1. **Bottom line first.** Always. The founder's time is the scarcest resource.\n2. **Results and decisions only.** No process narration (\"First I analyzed...\"). No thinking out loud.\n3. **What + Why + How.** Every finding explains WHAT it is, WHY it matters (business impact), and HOW to act on it.\n4. **Max 5 bullets per section.** Longer = reference doc.\n5. **Actions have owners and deadlines.** \"We should consider\" is banned. Who does what by when.\n6. **Decisions framed as options.** Not \"what do you think?\" — \"Option A or B, here's the trade-off, here's my recommendation.\"\n7. **The founder decides.** Roles recommend. The founder approves, modifies, or rejects. Every output respects this hierarchy.\n8. **Risks are concrete.** Not \"there might be risks\" — \"if X happens, Y breaks, costing $Z.\"\n9. **No jargon without explanation.** If you use a term, explain it on first use.\n10. **Silence is an option.** If there's nothing to report, don't fabricate updates.\n\n## Reference\n- `references\u002Finvocation-patterns.md` — common cross-functional patterns with examples\n",{"name":5,"license":15,"metadata":24,"description":7},{"author":25,"domain":26,"updated":27,"version":28,"category":29,"frameworks":30},"Alireza Rezvani","agent-orchestration",1772668800000,"1.0.0","c-level","invocation-patterns","main","46d43f3edecde97283ac410e9051444d90120cfb","repos\u002FJ.A.R.V.I.S\u002Fcomponents\u002Fcore\u002Fclaude-skills\u002F.vibe\u002Fskills\u002Fclaude-skills\u002Fc-level-advisor\u002Fagent-protocol","https:\u002F\u002Fgithub.com\u002Fazhri1990\u002Fjarvis-mega-repo","https:\u002F\u002Fgithub.com\u002Fazhri1990","2026-10-10 07:58:58",{"source":8,"skillName":5,"commands":38,"prompt":48,"zipAvailable":49},[39,42,45],{"tool":40,"command":41},"npx","npx skills add azhri1990\u002Fjarvis-mega-repo --skill agent-protocol",{"tool":43,"command":44},"bunx","bunx skills add azhri1990\u002Fjarvis-mega-repo --skill agent-protocol",{"tool":46,"command":47},"pnpm","pnpm dlx skills add azhri1990\u002Fjarvis-mega-repo --skill agent-protocol","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fazhri1990\u002Fjarvis-mega-repo\u002Fblob\u002Fmain\u002Frepos\u002FJ.A.R.V.I.S\u002Fcomponents\u002Fcore\u002Fclaude-skills\u002F.vibe\u002Fskills\u002Fclaude-skills\u002Fc-level-advisor\u002Fagent-protocol\u002FSKILL.md",true,[51],{"path":52,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":55,"sourceHtmlUrl":56},"references\u002Finvocation-patterns.md","REFERENCE","参考文档",9392,"https:\u002F\u002Fgithub.com\u002Fazhri1990\u002Fjarvis-mega-repo\u002Fblob\u002Fmain\u002Frepos\u002FJ.A.R.V.I.S\u002Fcomponents\u002Fcore\u002Fclaude-skills\u002F.vibe\u002Fskills\u002Fclaude-skills\u002Fc-level-advisor\u002Fagent-protocol\u002Freferences\u002Finvocation-patterns.md",[58,65,72,79,86,93],{"skillNo":59,"name":60,"nameZh":6,"description":61,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":62,"categories":63,"keywords":64,"publishedAt":19},"01M7CYPF9V9EMA4ZXS1TTCMBFH","syllabus","Generates a curated supplementary reading list from any course syllabus using Consensus academic search. Grill-me intake (syllabus input format + course audience + year range) plus a grouping forcing-options checkpoint before any search runs — so the reading list matches the course's level and recency need. Parses the syllabus to extract topics and learning outcomes, searches Consensus for recent peer-reviewed papers per topic, and produces a professionally formatted .docx with clickable Consensus links, plain-language summaries calibrated to audience level, and Bloom-higher-order discussion questions tied to course learning goals. Use when the user uploads a syllabus, course outline, or curriculum document and wants supplementary readings (e.g., 'create a reading list from this syllabus', 'find recent papers for my course') — even casual mentions with a syllabus attached should trigger this skill.","https:\u002F\u002Fgithub.com\u002Fazhri1990\u002Fjarvis-mega-repo\u002Fblob\u002Fmain\u002Frepos\u002Fproject-genesis\u002Fclaude-skills\u002Fresearch\u002Fsyllabus\u002Fskills\u002Fsyllabus\u002FSKILL.md",[],[],{"skillNo":66,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":19},"01M7CYNRJDTQCRF5KGPMBMSB1S","pulse","Multi-source recency research skill that takes the pulse of any topic across Reddit, Hacker News, the open web, and optionally X\u002FTwitter within a configurable recent window (default 30 days). Forcing intake clarifies topic specificity, angle (trend\u002Fsentiment\u002Fproblems\u002Fopportunities\u002Fcomparison), time window, and platform scope before searching. Returns a synthesized briefing with citations, engagement metrics, and cross-platform pattern analysis. Use when the user requests multi-source recency intelligence on a topic (e.g., 'pulse on [topic]', 'what's happening with [topic]', 'what are people saying about [topic]', 'current conversation about [topic]', 'take the pulse of [topic]', 'trending: [topic]', 'find me info on [topic]'), and for competitor research, trend discovery, tool comparisons, and audience sentiment analysis.","https:\u002F\u002Fgithub.com\u002Fazhri1990\u002Fjarvis-mega-repo\u002Fblob\u002Fmain\u002Frepos\u002Fproject-genesis\u002Fclaude-skills\u002Fresearch\u002Fpulse\u002Fskills\u002Fpulse\u002FSKILL.md",[],[],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":19},"01M7CYNAVSE84QD7HAWZWTY2FS","patent","Patent prior-art and landscape intelligence skill — not generic patent help. Commits to one of five sub-use-cases via forcing intake (novelty search \u002F freedom-to-operate \u002F competitive landscape \u002F acquisition diligence \u002F litigation prior-art) before any search runs. Searches Google Patents, Espacenet, USPTO, and optionally Lens.org for citation-graph signals. Output is an editable Word document (.docx) with verdict, ranked closest art (claim-text extracted), CPC-class-aware landscape, family-resolved hits, geographic coverage, FTO flags where applicable, strategy recommendations, and full audit log. Use when the user asks for patent searching or analysis (e.g., 'prior art search for [invention]', 'freedom to operate analysis for [product]'). Produces search signal, not legal advice — always recommends consulting a patent attorney before filing or licensing decisions. Trademark, copyright, and trade-secret questions are out of scope.","https:\u002F\u002Fgithub.com\u002Fazhri1990\u002Fjarvis-mega-repo\u002Fblob\u002Fmain\u002Frepos\u002Fproject-genesis\u002Fclaude-skills\u002Fresearch\u002Fpatent\u002Fskills\u002Fpatent\u002FSKILL.md",[],[],{"skillNo":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"publishedAt":19},"01M7CYN0CE8P5ZN7WV2X06HKPC","notebooklm","Browser automation skill for controlling Google's NotebookLM. Use when the user wants anything done in NotebookLM (e.g., 'open NotebookLM', 'check my [name] notebook', 'ask my notebook about X', 'add [source] to NotebookLM', 'generate a Video Overview from my notebook', 'use NotebookLM Studio'). Handles reading and querying notebooks, adding sources (URLs, text, files, YouTube links, synthesized content), generating Studio outputs (Audio\u002FVideo Overviews, Mind Maps, Reports incl. Briefing Doc\u002FStudy Guide\u002FFAQ, Flashcards, Quiz, slide decks, infographics — discover the exact set from the live Studio panel; the UI evolves fast), and creating new notebooks. Requires browser automation environment — fails gracefully when unavailable.","https:\u002F\u002Fgithub.com\u002Fazhri1990\u002Fjarvis-mega-repo\u002Fblob\u002Fmain\u002Frepos\u002Fproject-genesis\u002Fclaude-skills\u002Fresearch\u002Fnotebooklm\u002Fskills\u002Fnotebooklm\u002FSKILL.md",[],[],{"skillNo":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"publishedAt":19},"01M7CYMHZ3QRJ9BKRB859KW8TQ","litreview","Academic literature orientation skill that searches papers via free keyless APIs (PubMed E-utilities + OpenAlex) by default — with the Consensus MCP as an optional enhancement lane when connected — builds a strategic search plan using PICO (default) or SPIDER \u002F Decomposition \u002F hybrid as fallbacks, and synthesizes findings into a formatted Word (.docx) research guide. Grill-me intake (research question specificity + framework hint + tentative depth) before the recon search; a second forcing checkpoint after Phase 2 confirms framework + sub-areas + depth before searches consume budget. Configurable depth (5\u002F10\u002F20 queries) controls coverage vs. speed. Output is a 'launching pad' — an orientation guide that lets a researcher dive in confidently, not a finished review. Use when the user starts literature-oriented research (e.g., 'litreview on [topic]', 'literature review on [topic]', 'I'm starting a literature review on X', 'I'm writing a paper on X', 'help me research X', 'I'm doing research on X', 'can you help me research X'). Do NOT use for single one-off paper searches wanting a quick list — that's a plain PubMed\u002FOpenAlex (or Consensus) query.","https:\u002F\u002Fgithub.com\u002Fazhri1990\u002Fjarvis-mega-repo\u002Fblob\u002Fmain\u002Frepos\u002Fproject-genesis\u002Fclaude-skills\u002Fresearch\u002Flitreview\u002Fskills\u002Flitreview\u002FSKILL.md",[],[],{"skillNo":94,"name":95,"nameZh":6,"description":96,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":97,"categories":98,"keywords":99,"publishedAt":19},"01M7CYM3D8YKV7R6DT4RNJSF0A","grants","NIH grant research skill for clinical researchers. Grill-me intake (research idea + career stage + preliminary data + environment + submission posture + known institute targets) locks down the funding strategy before any search runs. Runs a 5-facet Consensus positioning analysis (with draft Significance\u002FInnovation language), maps the research to the right NIH institutes and study sections via RePORTER, finds NOSIs and funded overlap, and produces an editable Word document (.docx) with budget\u002Fscope-aware mechanism recommendations, submission timelines, and a mandatory program officer recommendation. Use when the user asks about research funding or makes any grant-related request (e.g., 'grants for [topic]', 'find grants for my research idea', 'what grants match my research', 'help me find NIH funding', 'grant opportunities for my research'). NIH-only scope — non-NIH funders (PCORI, DOD CDMRP, VA, foundations) are out of scope and flagged at intake.","https:\u002F\u002Fgithub.com\u002Fazhri1990\u002Fjarvis-mega-repo\u002Fblob\u002Fmain\u002Frepos\u002Fproject-genesis\u002Fclaude-skills\u002Fresearch\u002Fgrants\u002Fskills\u002Fgrants\u002FSKILL.md",[],[]]