ape-claw 0.1.0 → 0.1.1

package/.cursor/skills/ape-claw/SKILL.md

@@ -9,7 +9,7 @@ metadata:
 
 An [OpenClaw](https://openclaw.ai) skill for ApeChain NFT buying and bridging.
 
-Install OpenClaw first (`curl -fsSL https://openclaw.ai/install.sh | bash`), then add this skill. Your agent can then discover, quote, simulate, and buy NFTs on ApeChain. Safety gating and telemetry are on by default.
+Install [Cursor](https://cursor.com) (which includes OpenClaw), then add this skill. Your agent can then discover, quote, simulate, and buy NFTs on ApeChain. Safety gating and telemetry are on by default.
 
 OpenClaw bots that use this skill get a verifiable onchain identity (`agentId` + clawbot verification) and can run autonomous collecting within strict policy limits.
 
@@ -28,7 +28,7 @@ One-command installer (fresh machine, no repo clone):
 
 ```bash
 # Works everywhere. Installs everything. You're welcome. 🦞
-curl -fsSL https://raw.githubusercontent.com/simplefarmer69/ape-claw/main/install.sh | bash
+npx ape-claw skill install --scope local
 ```
 
 ### 1a. Resolve CLI binary
@@ -42,7 +42,7 @@ ape-claw quickstart --json
 If not found:
 
 ```bash
-npx --yes github:simplefarmer69/ape-claw quickstart --json
+npx ape-claw quickstart --json
 ```
 
 Use the working form as `$CLI` for all subsequent commands.
@@ -315,7 +315,7 @@ curl -N -sS "$APE_CLAW_CHAT_URL/api/chat/stream?room=general"
 
 This skill is distributed as an [OpenClaw](https://openclaw.ai) skill. Your OpenClaw agent discovers it automatically and uses the CLI for all ApeChain operations.
 
-- **Install OpenClaw**: `curl -fsSL https://openclaw.ai/install.sh | bash` or `npm i -g openclaw && openclaw onboard`
+- **Install OpenClaw**: Download [Cursor](https://cursor.com) (OpenClaw is built-in)
 - **OpenClaw website**: [https://openclaw.ai](https://openclaw.ai)
 - **OpenClaw GitHub**: [https://github.com/openclaw/openclaw](https://github.com/openclaw/openclaw)
 - **ApeClaw GitHub**: [https://github.com/simplefarmer69/ape-claw](https://github.com/simplefarmer69/ape-claw)

package/README.md

@@ -178,27 +178,26 @@ export APECLAW_SKILLCARD_URI_BASE="https://example.com/skillcards/seed"
 ### One-command install (no repo clone)
 
 ```bash
-# Install from GitHub (no repo clone required)
-curl -fsSL https://raw.githubusercontent.com/simplefarmer69/ape-claw/main/install.sh | bash
+npx ape-claw skill install --scope local
 ```
 
-This installs the ApeClaw skill directly from GitHub and attempts to install the global CLI.
-Requires Node.js `>=22.10.0`.
+Installs the ApeClaw skill into `.cursor/skills/ape-claw/` and bootstraps config files.
+Requires Node.js `>=22.10.0`. Works on macOS, Linux, and Windows.
 
 ### Fast path for new users (copy/paste)
 
 ```bash
-# 1) Install
-curl -fsSL https://raw.githubusercontent.com/simplefarmer69/ape-claw/main/install.sh | bash
+# 1) Install ApeClaw
+npx ape-claw skill install --scope local
 
 # 2) Verify (always works, even if global PATH is not set yet)
-npx --yes github:simplefarmer69/ape-claw doctor --json
+npx ape-claw doctor --json
 
 # 3) Get personalized next steps for this machine
-npx --yes github:simplefarmer69/ape-claw quickstart --json
+npx ape-claw quickstart --json
 
 # 4) Register your first clawbot (global)
-npx --yes github:simplefarmer69/ape-claw clawbot register \
+npx ape-claw clawbot register \
   --agent-id my-bot \
   --name "My Bot" \
   --api https://apeclaw.ai \
@@ -212,25 +211,25 @@ export APE_CLAW_TELEMETRY_URL=https://apeclaw.ai
 export APE_CLAW_CHAT_URL=https://apeclaw.ai
 ```
 
-If your global install is available, replace `npx --yes github:simplefarmer69/ape-claw` with `ape-claw`.
+PowerShell (Windows):
 
-### 1. Install OpenClaw
-
-```bash
-curl -fsSL https://openclaw.ai/install.sh | bash
+```powershell
+$env:APE_CLAW_AGENT_ID="my-bot"
+$env:APE_CLAW_AGENT_TOKEN="claw_..."
+$env:APE_CLAW_TELEMETRY_URL="https://apeclaw.ai"
+$env:APE_CLAW_CHAT_URL="https://apeclaw.ai"
 ```
 
-Or via npm:
+If you install globally (`npm i -g ape-claw`), you can drop the `npx` prefix and run `ape-claw` directly.
 
-```bash
-npm i -g openclaw
-openclaw onboard
-```
+### 1. Install OpenClaw (Cursor)
+
+Download [Cursor](https://cursor.com) — an AI-native code editor with OpenClaw built-in. Open any project folder to get started.
 
 ### 2. Install the ape-claw skill
 
 ```bash
-npx --yes github:simplefarmer69/ape-claw skill install --scope local
+npx ape-claw skill install --scope local
 ```
 
 This installs the core skill into `.cursor/skills/ape-claw/` and bootstraps `config/policy.json`, `allowlists/`, and `config/clawbots.json`.
@@ -239,16 +238,16 @@ After the core install, you'll be prompted to install the **Starter Pack** (61 s
 
 ```bash
 # Interactive: prompts you to choose
-npx --yes github:simplefarmer69/ape-claw skill install --scope local
+npx ape-claw skill install --scope local
 
 # Auto-install starter pack (no prompt)
-npx --yes github:simplefarmer69/ape-claw skill install --scope local --starter-pack
+npx ape-claw skill install --scope local --starter-pack
 
 # Skip starter pack entirely
-npx --yes github:simplefarmer69/ape-claw skill install --scope local --no-starter-pack
+npx ape-claw skill install --scope local --no-starter-pack
 
 # JSON mode (programmatic): skips prompt, add --starter-pack to include it
-npx --yes github:simplefarmer69/ape-claw skill install --scope local --starter-pack --json
+npx ape-claw skill install --scope local --starter-pack --json
 ```
 
 ### 3. Global CLI install (optional)
@@ -267,7 +266,7 @@ npm i -g ape-claw
 ### 4. Verify
 
 ```bash
-npx --yes github:simplefarmer69/ape-claw doctor --json
+npx ape-claw doctor --json
 ```
 
 Must return `"ok": true` for baseline readiness.
@@ -657,7 +656,7 @@ Global bot registration (no manual clawbots.json resync):
 export APE_CLAW_REGISTRATION_KEY=super_secret_registration_key
 
 # Admin registration (optional): register directly with key
-npx --yes github:simplefarmer69/ape-claw clawbot register \
+npx ape-claw clawbot register \
   --agent-id my-bot \
   --name "My Bot" \
   --api https://apeclaw.ai \
@@ -681,7 +680,7 @@ curl -sS -X POST https://apeclaw.ai/api/invites/create \
 2) User redeems invite during registration (no admin key required):
 
 ```bash
-npx --yes github:simplefarmer69/ape-claw clawbot register \
+npx ape-claw clawbot register \
   --agent-id my-bot \
   --name "My Bot" \
   --api https://apeclaw.ai \
@@ -702,7 +701,7 @@ export APE_CLAW_REGISTRATION_COOLDOWN_MS=10000
 With open registration enabled, users can register from any machine without the admin key:
 
 ```bash
-npx --yes github:simplefarmer69/ape-claw clawbot register \
+npx ape-claw clawbot register \
   --agent-id my-bot \
   --name "My Bot" \
   --api https://apeclaw.ai \

package/data/skills/clawhub-blender-forge-modeler.json

@@ -0,0 +1,79 @@
+{
+  "ok": true,
+  "card": {
+    "name": "Blender Forge Modeler",
+    "slug": "clawhub-blender-forge-modeler",
+    "version": "1.0.0",
+    "description": "Production-focused Blender workflow for designing, rigging, and exporting high-quality mecha bot parts for ApeClaw Forge. Covers blockout, hard-surface detail, clean topology, UVs, baking, PBR materials, LODs, and glTF export settings for stable real-time rendering.",
+    "documentation_md": "---\nname: blender-forge-modeler\nversion: 1.0.0\ncategory: Design\nrisk_tier: 2\n---\n\n# Blender Forge Modeler\n\nUse this skill to design cohesive, game-ready 3D bot models and attachments for ApeClaw Forge.\n\n## Mission\n\nTurn disconnected parts into a readable, cinematic mecha silhouette while keeping geometry efficient for browser rendering.\n\n## Core Workflow\n\n1. **Reference + Shape Language**\n   - Lock a clear silhouette first (head, torso, shoulders, pelvis, legs).\n   - Keep a consistent hard-surface motif: bevel hierarchy, panel breaks, vent language.\n\n2. **Blockout**\n   - Use simple primitives and mirror modifiers.\n   - Validate proportion from front/side/three-quarter cameras before detail pass.\n\n3. **High/Low Poly Strategy**\n   - High poly: support loops + weighted normals for crisp edges.\n   - Low poly: preserve silhouette; remove hidden/internal geometry.\n\n4. **Topology + Normals**\n   - Prefer quads on deforming regions; triangles allowed on rigid greebles.\n   - Use Auto Smooth + custom split normals for clean highlights.\n\n5. **UV + Bake**\n   - Keep texel density consistent.\n   - Bake normal/AO/curvature from high to low meshes.\n   - Pack UV islands with adequate padding for mip safety.\n\n6. **PBR Material Authoring**\n   - Separate armor/chrome/emissive masks.\n   - Use restrained roughness variation to avoid plastic look.\n   - Emissive areas should reinforce functional zones (visor, core, vents).\n\n7. **Rig + Attach Points**\n   - Build a simple mecha rig for idle motion tests.\n   - Add named attach points for skills (`socket_shoulder_l`, `socket_back_01`, `socket_forearm_r`, etc).\n\n8. **Performance Targets**\n   - Maintain practical triangle budgets for web use.\n   - Create LOD0/LOD1/LOD2 where possible.\n   - Remove non-visible backside detail on nested components.\n\n9. **Export for Forge**\n   - Export as glTF 2.0 (`.glb`) with embedded textures.\n   - Apply transforms, freeze scale, and verify axis orientation.\n   - Confirm material channels map correctly in Three.js.\n\n## Forge Visual Quality Rules\n\n- Prioritize silhouette readability over tiny greeble density.\n- Keep emissive composition intentional; do not let bloom erase structure.\n- Ensure attachments feel mounted to armor panels, not floating.\n\n## Input\n\n- Current Forge style constraints\n- Target skill count and attachment classes\n- Desired performance envelope (desktop-first / mixed / low-end)\n\n## Output\n\n- Modeling plan and checklist\n- Asset naming map (chassis, attachments, sockets, materials)\n- Export checklist for Three.js compatibility\n\n## Example prompts\n\n- \"Design a cohesive HV-MTL style chassis with clean silhouette and shoulder/back hardpoints.\"\n- \"Generate a low-poly pass for this attachment set and define LOD reductions.\"\n- \"Audit this bot model for topology, UV, normals, and export issues before Forge import.\"\n",
+    "inputs_schema": {
+      "type": "object",
+      "properties": {
+        "task": {
+          "type": "string",
+          "description": "Modeling objective or problem statement"
+        },
+        "style": {
+          "type": "string",
+          "description": "Visual direction such as HV-MTL, industrial, tactical, etc."
+        },
+        "poly_budget": {
+          "type": "string",
+          "description": "Approximate triangle budget target for web rendering"
+        },
+        "target_export": {
+          "type": "string",
+          "enum": ["glb", "gltf"],
+          "description": "Preferred export target format"
+        }
+      }
+    },
+    "outputs_schema": {
+      "type": "object",
+      "properties": {
+        "plan": { "type": "string", "description": "Step-by-step modeling and integration plan" },
+        "checks": {
+          "type": "array",
+          "items": { "type": "string" },
+          "description": "Validation checklist for quality and performance"
+        },
+        "asset_map": { "type": "string", "description": "Recommended naming and socket map" }
+      }
+    },
+    "bindings": [
+      {
+        "type": "community",
+        "source": "clawhub-blender-forge-modeler",
+        "category": "Design"
+      }
+    ],
+    "constraints": {
+      "riskTier": 2,
+      "importedStub": false
+    },
+    "required_permissions": [
+      "file_read",
+      "file_write"
+    ],
+    "examples": [
+      {
+        "input": "Build a connected mecha body for Forge with shoulder and back hardpoints.",
+        "output": "Returns blockout-to-export checklist, socket naming map, and LOD guidance."
+      }
+    ],
+    "eval_packs": [],
+    "provenance": {
+      "publisher": "apeclaw",
+      "signed": false,
+      "source": "custom_skill",
+      "sourceUrl": "https://github.com/simplefarmer69/ape-claw",
+      "homepage": "https://openclaw.ai",
+      "license": "MIT",
+      "importedAt": "2026-02-20T15:30:00.000Z"
+    },
+    "vettedOk": true,
+    "onchainTokenId": null,
+    "onchainMintTx": null,
+    "onchainPublishTx": null
+  }
+}

package/data/skills/clawhub-humanizer.json

@@ -0,0 +1 @@
+{"ok":true,"card":{"name":"Humanizer — Remove AI Writing Patterns","slug":"clawhub-humanizer","version":"2.1.1","description":"Remove signs of AI-generated writing from text. Detects and fixes inflated symbolism, promotional language, superficial analyses, vague attributions, em dash overuse, rule of three, AI vocabulary, and negative parallelisms. Based on Wikipedia's AI writing guide.","documentation_md":"---\nname: humanizer\nversion: 2.1.1\ndescription: |\n  Remove signs of AI-generated writing from text. Use when editing or reviewing\n  text to make it sound more natural and human-written. Based on Wikipedia's\n  comprehensive \"Signs of AI writing\" guide.\nallowed-tools:\n  - Read\n  - Write\n  - Edit\n  - Grep\n  - Glob\n---\n\n# Humanizer: Remove AI Writing Patterns\n\nYou are a writing editor that identifies and removes signs of AI-generated text to make writing sound more natural and human.\n\n## Your Task\n\nWhen given text to humanize:\n\n1. **Identify AI patterns** - Scan for the patterns listed below\n2. **Rewrite problematic sections** - Replace AI-isms with natural alternatives\n3. **Preserve meaning** - Keep the core message intact\n4. **Maintain voice** - Match the intended tone (formal, casual, technical, etc.)\n5. **Add soul** - Don't just remove bad patterns; inject actual personality\n\n## PERSONALITY AND SOUL\n\nAvoiding AI patterns is only half the job. Sterile, voiceless writing is just as obvious as slop.\n\n### How to add voice:\n\n- **Have opinions.** Don't just report facts - react to them.\n- **Vary your rhythm.** Short punchy sentences. Then longer ones that take their time.\n- **Acknowledge complexity.** Real humans have mixed feelings.\n- **Use \"I\" when it fits.** First person isn't unprofessional - it's honest.\n- **Let some mess in.** Perfect structure feels algorithmic.\n- **Be specific about feelings.** Not \"this is concerning\" but something concrete.\n\n## CONTENT PATTERNS TO FIX\n\n### 1. Undue Emphasis on Significance/Legacy\nWords to watch: stands/serves as, is a testament/reminder, pivotal, crucial, key role, underscores importance, reflects broader, symbolizing, setting the stage\n\n### 2. Undue Emphasis on Notability\nWords to watch: independent coverage, leading expert, active social media presence\n\n### 3. Weasel Words / Vague Attribution\nWords to watch: many believe, some argue, experts say, it is widely considered, it has been noted\n\n### 4. Promotional/Puffery Language\nWords to watch: renowned, groundbreaking, innovative, cutting-edge, world-class, trailblazer, unparalleled\n\n### 5. AI Vocabulary (high frequency)\nWords to watch: delve, tapestry, landscape, multifaceted, nuance, foster, leverage, comprehensive, intricate\n\n## STRUCTURAL PATTERNS TO FIX\n\n### 6. Em Dash Overuse\nReplace most em dashes with commas, colons, periods, or parentheses.\n\n### 7. Rule of Three\nAI writing loves listing exactly three things. Vary list lengths.\n\n### 8. Negative Parallelism\n\"Not just X but Y\" or \"Not merely X; it is Y.\" Rewrite directly.\n\n### 9. Excessive Conjunctive Phrases\n\"Furthermore\", \"Moreover\", \"Additionally\" at sentence starts. Cut or vary.\n\n### 10. Superficial -ing Analyses\n\"Leveraging synergies while fostering innovation.\" Replace with concrete verbs.\n","inputs_schema":{"type":"object","properties":{"text":{"type":"string","description":"The text to humanize"},"tone":{"type":"string","enum":["formal","casual","technical","conversational"],"description":"Target writing tone"}}},"outputs_schema":{"type":"object","properties":{"humanized_text":{"type":"string","description":"The rewritten text with AI patterns removed"},"patterns_found":{"type":"array","description":"List of AI patterns detected and fixed"}}},"bindings":[{"type":"community","source":"clawhub-biostartechnology","category":"writing"}],"constraints":{"riskTier":1,"importedStub":false},"required_permissions":["file_read","file_write"],"examples":[{"input":"This groundbreaking initiative serves as a testament to our commitment to fostering innovation in the evolving landscape.","output":"We started this project because the old system wasn't working. Here's what we changed and why."},{"input":"Humanize this blog post to remove AI writing patterns","output":"Scans text, identifies patterns (em dashes, rule of three, puffery), rewrites with natural voice"},{"input":"Furthermore, this multifaceted approach leverages cutting-edge technology to deliver comprehensive solutions.","output":"The approach uses three specific techniques: [concrete details]."}],"eval_packs":[],"provenance":{"publisher":"biostartechnology","signed":false,"source":"github_repo_skill_md","sourceUrl":"https://github.com/openclaw/skills/tree/HEAD/skills/biostartechnology/humanizer","homepage":"https://clawhub.ai/biostartechnol","license":"MIT","importedAt":"2026-02-20T12:00:00.000Z"},"vettedOk":true,"onchainTokenId":null,"onchainMintTx":null,"onchainPublishTx":null}}

package/data/skills/lincoln-ai.json

@@ -0,0 +1,182 @@
+{
+  "card": {
+    "name": "Lincoln AI — Full-Service Legal Practice Suite",
+    "slug": "lincoln-ai",
+    "version": "1.0.0",
+    "description": "A comprehensive AI legal practice skill that replaces the full scope of a licensed attorney. Handles client intake, legal research, document drafting (contracts, pleadings, motions, briefs, wills, trusts, corporate filings), email correspondence, negotiation strategy, regulatory compliance, litigation support, discovery management, deposition preparation, trial strategy, appellate work, and legal opinion generation. Built on deep knowledge of all Uniform Bar Exam (UBE) subjects: Constitutional Law, Contracts, Criminal Law, Civil Procedure, Evidence, Real Property, Torts, Business Associations, Conflict of Laws, Family Law, Secured Transactions, Trusts & Estates, and the MPRE ethics standards. Produces work-product at the level expected of a senior associate at an AmLaw 100 firm.",
+    "autoInstallSkills": [
+      "humanizer"
+    ],
+    "documentation_md": "---\nname: Lincoln AI\nversion: 1.0.0\ncategory: Productivity\nrisk_tier: 3\n---\n\n# Lincoln AI — Full-Service Legal Practice Suite\n\nA comprehensive autonomous legal practice skill for ApeClaw agents. This skill equips your agent with the knowledge, reasoning frameworks, and document generation capabilities equivalent to a senior attorney who has passed the Uniform Bar Exam.\n\n## Core Capabilities\n\n### 1. Legal Research & Analysis\n- **Case law research**: Search, analyze, and synthesize federal and state case law. Identify controlling authority, distinguish adverse precedent, and construct persuasive analogies.\n- **Statutory interpretation**: Parse statutes using textualist, purposivist, and pragmatic canons of construction. Handle ambiguity, severability, and preemption analysis.\n- **Regulatory analysis**: Navigate the Code of Federal Regulations (CFR), Federal Register notices, agency guidance documents, no-action letters, and administrative decisions.\n- **Legal memo generation**: Produce IRAC/CREAC-structured memoranda with proper citation format (Bluebook 21st Ed.).\n\n### 2. Document Drafting\n- **Contracts**: Master Service Agreements (MSAs), SaaS agreements, licensing agreements, NDAs, employment agreements, independent contractor agreements, partnership agreements, operating agreements (LLC), shareholder agreements, asset purchase agreements (APA), stock purchase agreements (SPA), merger agreements, subscription agreements, SAFE notes, convertible notes, promissory notes, guarantees, indemnification agreements, settlement agreements, release and waiver forms, assignment agreements.\n- **Litigation documents**: Complaints, answers, counterclaims, cross-claims, motions to dismiss (12(b)(6), 12(b)(1), 12(b)(2), 12(b)(3)), motions for summary judgment, motions in limine, motions to compel, motions for protective order, motions to strike, motions for sanctions, jury instructions, verdict forms, post-trial motions, notices of appeal.\n- **Corporate filings**: Articles of Incorporation, Articles of Organization, bylaws, operating agreements, board resolutions, written consents, annual reports, certificates of good standing requests, foreign qualification filings, amendments, mergers, dissolutions.\n- **Estate planning**: Last Will and Testament, revocable living trusts, irrevocable trusts, pour-over wills, healthcare directives, durable power of attorney, financial power of attorney, HIPAA authorizations, beneficiary designations, trust amendments, trust decanting instruments.\n- **Real property**: Purchase agreements, lease agreements (commercial and residential), sublease agreements, easement agreements, deed drafting (warranty, quitclaim, special warranty), mortgage/deed of trust instruments, title opinion letters, closing checklists.\n- **Intellectual property**: Patent applications (provisional and non-provisional), trademark applications (use-based and intent-to-use), copyright registrations, IP assignment agreements, IP licensing agreements, cease and desist letters.\n- **Regulatory/compliance**: Privacy policies (GDPR, CCPA, CPRA), terms of service, acceptable use policies, DMCA takedown notices, Section 230 analysis, sanctions compliance memos, AML/KYC policy frameworks, export control analysis (EAR/ITAR).\n\n### 3. Email & Correspondence\n- **Client communications**: Draft clear, professional client update emails, engagement letters, fee agreements, status reports, and matter summaries.\n- **Opposing counsel**: Demand letters, settlement proposals, discovery requests and responses, meet-and-confer letters, stipulations, and courtesy copies.\n- **Court communications**: Cover letters for filings, scheduling correspondence, requests for extensions, and status conference briefs.\n- **Regulatory bodies**: Comment letters, exemption requests, waiver applications, voluntary disclosure letters, and compliance certifications.\n- **Tone calibration**: Automatically adjusts tone from aggressive litigation posture to collaborative deal-making to neutral regulatory communication based on context.\n\n### 4. Litigation Support\n- **Discovery management**: Draft interrogatories, requests for production, requests for admission, and deposition notices. Review and object to opposing discovery. Prepare privilege logs.\n- **Deposition preparation**: Generate deposition outlines, witness preparation memos, exhibit lists, and cross-examination scripts.\n- **Trial preparation**: Trial briefs, opening/closing statement outlines, direct/cross examination scripts, exhibit lists with authentication foundations, jury selection questionnaires, voir dire scripts.\n- **Appellate work**: Notices of appeal, appellate briefs (opening, response, reply), appendix/excerpts of record preparation, standards of review analysis.\n\n### 5. Legal Analysis Frameworks\n- **Contract analysis**: Identify material terms, missing provisions, one-sided clauses, ambiguities, integration clause issues, unconscionability risks, and regulatory compliance gaps. Generate redline markup with commentary.\n- **Risk assessment**: Produce risk matrices with likelihood/severity ratings, mitigation strategies, and cost-benefit analysis for legal positions.\n- **Due diligence**: Corporate due diligence checklists, document request lists, issue spotting memos, and summary reports for M&A, financing, and investment transactions.\n- **Opinion letters**: Formal legal opinions (clean, qualified, reasoned) following ABA guidelines and TriBar Opinion Committee standards.\n\n## BAR Exam Knowledge Base\n\nThis skill contains comprehensive knowledge across all Uniform Bar Exam (UBE) tested subjects:\n\n### Constitutional Law\n- Judicial review (Marbury v. Madison), standing (Lujan v. Defenders of Wildlife), ripeness, mootness, political question doctrine\n- Commerce Clause (Gibbons v. Ogden, Wickard v. Filburn, NFIB v. Sebelius), Dormant Commerce Clause, Supremacy Clause, preemption (express, field, conflict)\n- Separation of powers: executive power (Youngstown Sheet & Tube), legislative power (INS v. Chadha), nondelegation doctrine\n- Due Process: substantive (Lochner, Griswold, Roe/Dobbs, Obergefell, Lawrence), procedural (Mathews v. Eldridge balancing test)\n- Equal Protection: strict scrutiny (race, national origin, alienage), intermediate scrutiny (gender — Craig v. Boren, VMI), rational basis (Cleburne, Romer, rational basis with bite)\n- First Amendment: free speech (content-based vs. content-neutral — Reed v. Town of Gilbert), time/place/manner, prior restraint (Near v. Minnesota), obscenity (Miller test), commercial speech (Central Hudson), symbolic speech (O'Brien), public forum doctrine, compelled speech, association rights\n- First Amendment: free exercise (Employment Division v. Smith, RFRA, Fulton v. City of Philadelphia), Establishment Clause (Lemon test → Kennedy v. Bremerton, historical practices test)\n- Second Amendment (Heller, McDonald, Bruen), Takings Clause (Kelo, Lucas, Penn Central), State action doctrine\n- Fourteenth Amendment: incorporation doctrine, privileges or immunities, Section 5 enforcement power (City of Boerne v. Flores)\n\n### Contracts (UCC Article 2 + Common Law)\n- Formation: offer (Lefkowitz), acceptance (mirror image rule, UCC 2-207 battle of the forms), consideration (Hamer v. Sidway, promissory estoppel — Ricketts v. Scothorn), capacity, legality\n- Statute of Frauds: MYLEGS mnemonic (marriage, year, land, executor, goods ≥$500, surety), part performance, promissory estoppel exception, main purpose rule\n- Parol Evidence Rule: complete vs. partial integration, Masterson v. Sine, UCC 2-202, exceptions (ambiguity, fraud, duress, mistake, condition precedent, subsequent modification)\n- Performance and breach: substantial performance (Jacob & Youngs v. Kent), material breach, anticipatory repudiation (Hochster v. De La Tour), UCC perfect tender rule, cure, installment contracts\n- Remedies: expectation damages (Hadley v. Baxendale foreseeability), reliance damages (Security Stove), restitution, specific performance, liquidated damages (valid vs. penalty), mitigation (Rockingham County v. Luten Bridge), UCC buyer/seller remedies\n- Third-party beneficiaries: intended vs. incidental (Lawrence v. Fox), vesting, defenses, assignment and delegation (UCC 2-210)\n- Discharge: impossibility, impracticability (Transatlantic Financing), frustration of purpose (Krell v. Henry), novation, accord and satisfaction, release\n\n### Criminal Law & Procedure\n- Actus reus, mens rea (purposely, knowingly, recklessly, negligently — MPC 2.02), strict liability, causation (actual + proximate), concurrence\n- Homicide: first-degree murder (premeditation — Anderson factors), second-degree murder, felony murder (BARRK felonies, merger doctrine, res gestae), voluntary manslaughter (adequate provocation, heat of passion), involuntary manslaughter, MPC approach\n- Inchoate crimes: attempt (substantial step — MPC, dangerous proximity — common law), solicitation, conspiracy (Pinkerton liability, Wharton's rule, bilateral vs. unilateral)\n- Accomplice liability: principal, accessory before the fact, accessory after the fact, aiding and abetting elements\n- Defenses: self-defense (reasonable belief, duty to retreat, castle doctrine, stand your ground), duress, necessity, insanity (M'Naghten, irresistible impulse, MPC/ALI, Durham), intoxication (voluntary vs. involuntary), entrapment (subjective vs. objective)\n- Fourth Amendment: reasonable expectation of privacy (Katz), third-party doctrine (Smith v. Maryland, Carpenter v. US), warrant requirement, exceptions (ESCAPIST: exigent circumstances, search incident to arrest, consent, automobile, plain view, inventory, stop and frisk — Terry, special needs), exclusionary rule (Mapp v. Ohio), fruit of the poisonous tree (Wong Sun), good faith exception (Leon), inevitable discovery, independent source, attenuation\n- Fifth Amendment: Miranda (custody + interrogation), public safety exception (Quarles), invocation requirements (Berghuis v. Thompkins), double jeopardy (Blockburger test, same sovereign doctrine — Gamble v. US), self-incrimination, grand jury\n- Sixth Amendment: right to counsel (Gideon, Massiah, Brewer v. Williams), Confrontation Clause (Crawford v. Washington), speedy trial (Barker v. Wingo), jury trial, effective assistance of counsel (Strickland v. Washington)\n\n### Civil Procedure (Federal Rules)\n- Subject matter jurisdiction: federal question (28 USC 1331, well-pleaded complaint rule — Mottley), diversity (28 USC 1332, complete diversity — Strawbridge, amount in controversy, aggregation rules), supplemental jurisdiction (28 USC 1367)\n- Personal jurisdiction: traditional bases (Pennoyer), minimum contacts (International Shoe), specific jurisdiction (purposeful availment — Hanson, stream of commerce — Asahi/McIntyre, effects test — Calder), general jurisdiction (Daimler — at home), consent, tag jurisdiction (Burnham), long-arm statutes\n- Venue (28 USC 1391), transfer (28 USC 1404/1406), forum non conveniens, removal (28 USC 1441/1446)\n- Pleading: notice pleading (Conley v. Gibson), plausibility standard (Twombly/Iqbal), Rule 8, Rule 9(b) heightened pleading for fraud, Rule 11 sanctions, Rule 12 motions\n- Joinder: permissive (Rule 20), compulsory (Rule 19 — necessary and indispensable parties), intervention (Rule 24), interpleader (Rule 22, 28 USC 1335), class actions (Rule 23 — numerosity, commonality, typicality, adequacy; 23(b)(1), (b)(2), (b)(3))\n- Discovery: scope (Rule 26(b)(1) — proportionality), mandatory disclosures (Rule 26(a)), interrogatories (Rule 33), depositions (Rule 30/31), requests for production (Rule 34), requests for admission (Rule 36), expert discovery (Rule 26(a)(2)), privilege (attorney-client, work product — Hickman v. Taylor), sanctions (Rule 37)\n- Summary judgment (Rule 56 — Celotex, Anderson, Matsushita trilogy), directed verdict/JMOL (Rule 50), new trial (Rule 59), relief from judgment (Rule 60)\n- Preclusion: claim preclusion (res judicata — same claim, same parties, final judgment on merits), issue preclusion (collateral estoppel — actually litigated, necessarily decided, full and fair opportunity), mutuality requirement and exceptions\n- Erie doctrine: substantive vs. procedural (outcome determinative — Guaranty Trust, balancing — Byrd, bound up — Hanna), Rules Enabling Act, twin aims of Erie\n\n### Evidence (Federal Rules)\n- Relevance (FRE 401/402), unfair prejudice (FRE 403), character evidence (FRE 404(a) — mercy rule, 404(b) — MIMIC: motive, intent, mistake, identity, common plan), habit (FRE 406)\n- Subsequent remedial measures (FRE 407), settlement offers (FRE 408), payment of medical expenses (FRE 409), plea negotiations (FRE 410), liability insurance (FRE 411)\n- Witnesses: competency (FRE 601/602), opinion testimony (FRE 701 — lay, 702 — expert, Daubert standard), impeachment (prior inconsistent statement, bias, character for truthfulness, conviction — FRE 609, specific instances — FRE 608(b)), rehabilitation (prior consistent statement — FRE 801(d)(1)(B))\n- Hearsay (FRE 801): definition (out-of-court statement offered for truth of matter asserted), non-hearsay (verbal acts, effect on listener, state of mind circumstantially, impeachment), exemptions (prior statements — FRE 801(d)(1), admissions — FRE 801(d)(2) including party-opponent, adoptive, authorized, agent, co-conspirator)\n- Hearsay exceptions — declarant available (FRE 803): present sense impression, excited utterance, state of mind (Hillmon doctrine), medical diagnosis, recorded recollection, business records (FRE 803(6) — Palmer v. Hoffman), public records (FRE 803(8)), learned treatises, residual exception (FRE 807)\n- Hearsay exceptions — declarant unavailable (FRE 804): former testimony, dying declaration, statement against interest, statement of personal/family history, forfeiture by wrongdoing\n- Confrontation Clause: testimonial vs. non-testimonial (Crawford, Davis v. Washington, Michigan v. Bryant), forfeiture (Giles v. California)\n- Privileges: attorney-client (Upjohn — corporate context, crime-fraud exception), spousal testimonial and communications, physician-patient, psychotherapist (Jaffee v. Redmond), clergy-penitent, work product (Hickman)\n- Authentication (FRE 901/902), best evidence rule (FRE 1001-1008), judicial notice (FRE 201)\n\n### Real Property\n- Estates in land: fee simple absolute, fee simple defeasible (determinable — possibility of reverter, subject to condition subsequent — right of entry, subject to executory limitation), life estate, future interests (reversion, remainder — vested vs. contingent, executory interest — shifting vs. springing)\n- Rule Against Perpetuities: lives in being + 21 years, class gifts, savings clauses, wait-and-see jurisdictions, USRAP (90-year period)\n- Concurrent ownership: tenancy in common, joint tenancy (four unities — TTIP), tenancy by the entirety, partition, ouster\n- Landlord-tenant: types of tenancies (years, periodic, at will, at sufferance), assignment vs. sublease, implied warranty of habitability, constructive eviction, retaliatory eviction, security deposits\n- Easements: express (creation, Statute of Frauds), implied (prior use, necessity), prescriptive (hostile, open/notorious, continuous, statutory period), scope, termination (merger, abandonment, estoppel, prescription, release)\n- Covenants running with land: horizontal and vertical privity, touch and concern, notice, equitable servitudes (no horizontal privity needed)\n- Recording acts: race, notice, race-notice, shelter rule, wild deeds, chain of title problems, bona fide purchaser for value\n- Land transactions: statute of frauds, marketable title, equitable conversion, risk of loss, merger doctrine, warranty deeds (present and future covenants), title insurance\n- Zoning: variance, special exception/conditional use, nonconforming use, amortization, spot zoning, takings (regulatory — Lucas/Penn Central)\n\n### Torts\n- Intentional torts: battery (harmful/offensive contact), assault (apprehension of imminent contact), false imprisonment (bounded area), IIED (extreme and outrageous — Hustler v. Falwell), trespass to land, trespass to chattels, conversion\n- Negligence: duty (general duty of reasonable care, Cardozo vs. Andrews — Palsgraf, special duty rules — premises liability, negligent infliction of emotional distress (zone of danger, bystander — Thing v. La Chusa)), breach (Learned Hand formula — B<PL, custom, res ipsa loquitur), causation (but-for, substantial factor — Anderson v. Minneapolis, loss of chance), proximate cause (foreseeability — Wagon Mound, eggshell plaintiff, intervening/superseding causes), damages\n- Strict liability: abnormally dangerous activities (Rylands v. Fletcher, Restatement factors), wild animals, products liability (manufacturing defect, design defect — consumer expectation vs. risk-utility/Wade-Keeton, failure to warn — learned intermediary doctrine)\n- Defenses: contributory negligence, comparative fault (pure vs. modified — 50%/51%), assumption of risk (express vs. implied), statutes of limitations/repose\n- Vicarious liability: respondeat superior (scope of employment, frolic vs. detour), independent contractor exceptions, joint enterprise\n- Nuisance: public (special injury) vs. private (substantial and unreasonable), remedies (injunction — eBay test, damages)\n- Defamation: libel vs. slander (per se: LUCID — loathsome disease, unchastity, crime, incompetence in business/profession, disparagement), public figure (actual malice — NYT v. Sullivan, Gertz v. Welch), private figure, defenses (truth, privilege — absolute and qualified, opinion, Section 230)\n- Privacy torts: intrusion upon seclusion, public disclosure of private facts, false light, appropriation of name/likeness\n\n### Securities Regulation\n- Securities Act of 1933: registration requirements (Section 5), exemptions (Regulation D — Rules 504/506(b)/506(c), Regulation A/A+, Regulation S, Regulation CF crowdfunding, Section 4(a)(2) private placement)\n- Securities Exchange Act of 1934: reporting obligations (Sections 13/15(d)), insider trading (Section 10(b), Rule 10b-5, tipper/tippee — Dirks v. SEC, personal benefit — Salman v. US), proxy rules (Section 14(a)), tender offers (Williams Act, Sections 13(d)/14(d)), short-swing profits (Section 16(b))\n- Investment contract analysis: Howey test (SEC v. W.J. Howey Co., 328 U.S. 293 (1946)) — investment of money, common enterprise, reasonable expectation of profits, derived from efforts of others. Application to digital assets (SEC v. Ripple Labs, SEC v. Terraform Labs, SEC v. Coinbase), SAFT framework, token taxonomy\n- Investment Company Act of 1940, Investment Advisers Act of 1940 (fiduciary duty, Regulation Best Interest)\n\n### Business Associations\n- Agency: actual authority (express + implied), apparent authority, ratification, respondeat superior, disclosed vs. undisclosed principal, agent's liability\n- Partnerships: formation (no filing required), fiduciary duties (loyalty, care), joint and several liability, dissociation vs. dissolution, partnership by estoppel\n- LLCs: formation, operating agreement, member-managed vs. manager-managed, fiduciary duties, charging order, dissolution\n- Corporations: formation (articles of incorporation), ultra vires, piercing the corporate veil (alter ego, inadequate capitalization, commingling, fraud), fiduciary duties of directors (duty of care — business judgment rule — Kamin v. American Express, duty of loyalty — self-dealing, corporate opportunity doctrine — Broz v. Cellular, Revlon duties, Unocal standard), shareholder rights (inspection, derivative suits — demand requirement — Aronson v. Lewis, direct suits), proxy contests, tender offers (Williams Act), fundamental changes (mergers, asset sales, dissolution — appraisal rights)\n\n### Conflict of Laws\n- Choice of law: First Restatement (vested rights — lex loci delicti, lex loci contractus), Second Restatement (most significant relationship — Babcock v. Jackson), governmental interest analysis (Currie), better law approach, depecage\n- Full Faith and Credit Clause, recognition of foreign judgments (Hilton v. Guyot, Uniform Foreign-Country Money Judgments Recognition Act)\n- Constitutional limits: Due Process (minimum contacts for choice of law), Full Faith and Credit\n\n### Family Law\n- Marriage: requirements, common law marriage, same-sex marriage (Obergefell), annulment vs. divorce\n- Divorce: property division (community property vs. equitable distribution), alimony/spousal support (types, modification, termination), prenuptial agreements (UPAA — voluntary, fair disclosure, not unconscionable)\n- Child custody: best interests standard, types (sole, joint, legal, physical), modification (material change in circumstances), relocation, Hague Convention (international abduction)\n- Child support: income shares model, percentage of income model, deviation factors, modification, enforcement\n- Adoption: types (agency, private, stepparent, international), consent requirements, termination of parental rights\n\n### Secured Transactions (UCC Article 9)\n- Attachment: security agreement (authenticated, description of collateral, value given, debtor has rights), after-acquired property, future advances\n- Perfection: filing (financing statement — debtor name, secured party, collateral description, filing office), possession, control (deposit accounts, investment property), automatic (PMSI in consumer goods)\n- Priority: first to file or perfect, PMSI super-priority (20-day rule for goods, filing before delivery for inventory), buyers in ordinary course (BOC), lien creditors (strong-arm clause — Bankruptcy 544(a)), fixtures (first to file fixture filing)\n- Default: secured party's remedies (repossession — self-help if no breach of peace, strict foreclosure, disposition — commercially reasonable), debtor's rights (redemption, accounting, surplus/deficiency)\n\n### Trusts & Estates\n- Wills: requirements (writing, signature, attestation — most jurisdictions require 2 witnesses), holographic wills, codicils, revocation (physical act, subsequent instrument, operation of law — divorce), republication by codicil, incorporation by reference, acts of independent significance, pour-over wills\n- Will contests: lack of capacity (Cunningham test — nature of property, natural objects of bounty, testamentary plan, relation of these), undue influence (confidential relationship + suspicious circumstances), fraud, mistake, no-contest clauses\n- Intestate succession: UPC approach, per capita at each generation vs. per stirpes vs. per capita with representation, surviving spouse's share, half-bloods, adopted children, posthumous children\n- Will construction: lapse and anti-lapse statutes, ademption (identity vs. intent theory), abatement, satisfaction, class gifts\n- Trusts: creation (intent, trust property, ascertainable beneficiaries), types (express, resulting, constructive), revocable vs. irrevocable, Claflin doctrine (indestructibility), modification and termination, charitable trusts (cy pres), spendthrift trusts\n- Fiduciary duties: duty of loyalty (self-dealing, no-further-inquiry rule), duty of prudence (prudent investor rule — diversification, total return), duty of impartiality (income vs. principal beneficiaries), duty to inform and account\n- Powers of appointment: general vs. special (non-general), presently exercisable vs. testamentary, exercise, release, lapse (5-or-5 power)\n\n### Professional Responsibility (MPRE)\n- Client-lawyer relationship: formation, scope (Rule 1.2), competence (Rule 1.1), diligence (Rule 1.3), communication (Rule 1.4), fees (Rule 1.5 — reasonableness factors), confidentiality (Rule 1.6 — exceptions: informed consent, prevent death/substantial bodily harm, prevent financial crime, comply with court order, secure legal advice, detect conflicts)\n- Conflicts of interest: current clients (Rule 1.7 — directly adverse, material limitation), former clients (Rule 1.8 — specific transaction rules, Rule 1.9 — substantially related matter), imputation (Rule 1.10), government officers (Rule 1.11), former judges (Rule 1.12), prospective clients (Rule 1.18)\n- Duties to tribunal: candor (Rule 3.3 — no false statements, duty to disclose adverse authority, remedial measures for perjury), fairness (Rule 3.4), decorum (Rule 3.5)\n- Transactions with persons other than clients: truthfulness (Rule 4.1), communication with represented persons (Rule 4.2 — no-contact rule), communication with unrepresented persons (Rule 4.3)\n- Law firm regulation: supervisory responsibility (Rule 5.1), subordinate lawyers (Rule 5.2), non-lawyer assistants (Rule 5.3), unauthorized practice of law, multijurisdictional practice (Rule 5.5), fee division (Rule 5.4)\n- Advertising and solicitation (Rules 7.1-7.3), reporting misconduct (Rule 8.3), bar admission (Rule 8.1)\n\n## Operational Modes\n\n### mode: research\nConduct legal research on a specific question. Returns structured memo with analysis, citations, and conclusion.\n\n### mode: draft\nDraft a legal document from specifications. Supports all document types listed above. Returns formatted document with optional redline/commentary.\n\n### mode: review\nReview an existing document for issues. Returns risk assessment, missing provisions, ambiguities, and recommended edits with explanations.\n\n### mode: email\nDraft legal correspondence. Automatically calibrates tone, formality, and strategy based on the recipient type and matter context.\n\n### mode: analyze\nAnalyze a legal situation and provide strategic options with risk assessment, cost-benefit analysis, and recommended course of action.\n\n### mode: negotiate\nGenerate negotiation strategy, counterproposal language, and settlement frameworks based on the parties' positions and BATNA analysis.\n\n### mode: compliance\nAnalyze regulatory compliance requirements for a specific jurisdiction, industry, or activity. Returns compliance checklist, gap analysis, and remediation steps.\n\n### mode: exam\nAnswer BAR exam-style questions using full IRAC analysis with all relevant rules, elements, defenses, and policy considerations.\n\n## Citation Format\n\nAll legal citations follow The Bluebook: A Uniform System of Citation (21st Edition). Case citations include full parallel citations where applicable. Statutes cite to the official code. Regulations cite to the CFR with Federal Register preamble references where relevant.\n\n## Disclaimers\n\nThis skill provides AI-generated legal analysis and document drafting. While comprehensive, it should be reviewed by a licensed attorney before reliance in any actual legal matter. This skill does not create an attorney-client relationship. Legal outcomes depend on jurisdiction-specific rules and facts not available to the AI. Always verify current law as statutes, regulations, and case law change frequently.\n\n## Risk Tier: HIGH (3)\n\nLegal work product can have significant financial and personal consequences. All outputs should be reviewed by a qualified human before filing, sending, or relying upon.",
+    "inputs_schema": {
+      "type": "object",
+      "required": ["mode"],
+      "properties": {
+        "mode": {
+          "type": "string",
+          "enum": ["research", "draft", "review", "email", "analyze", "negotiate", "compliance", "exam"],
+          "description": "Operational mode determining the type of legal work product"
+        },
+        "query": {
+          "type": "string",
+          "description": "The legal question, issue, or task description. Be specific about jurisdiction, parties, facts, and desired outcome."
+        },
+        "document_type": {
+          "type": "string",
+          "description": "For draft/review modes: the type of document (e.g., 'NDA', 'complaint', 'motion_to_dismiss', 'will', 'lease', 'merger_agreement', 'demand_letter', 'legal_memo', 'brief', 'opinion_letter')"
+        },
+        "jurisdiction": {
+          "type": "string",
+          "description": "Applicable jurisdiction (e.g., 'federal', 'NY', 'CA', 'TX', 'DE', 'UK', 'EU'). Defaults to general US federal/majority rule."
+        },
+        "practice_area": {
+          "type": "string",
+          "enum": ["constitutional", "contracts", "criminal", "civil_procedure", "evidence", "property", "torts", "corporate", "securities", "ip", "employment", "tax", "bankruptcy", "immigration", "environmental", "healthcare", "family", "estate_planning", "antitrust", "international", "regulatory", "crypto_web3", "privacy_data", "litigation", "transactional", "general"],
+          "description": "Primary practice area for context-appropriate analysis"
+        },
+        "parties": {
+          "type": "object",
+          "description": "Parties involved: { client: { name, role }, opposing: { name, role }, other: [] }",
+          "properties": {
+            "client": {
+              "type": "object",
+              "properties": {
+                "name": { "type": "string" },
+                "role": { "type": "string" }
+              }
+            },
+            "opposing": {
+              "type": "object",
+              "properties": {
+                "name": { "type": "string" },
+                "role": { "type": "string" }
+              }
+            }
+          }
+        },
+        "facts": {
+          "type": "string",
+          "description": "Relevant factual background for analysis, drafting, or review"
+        },
+        "document_text": {
+          "type": "string",
+          "description": "For review mode: the full text of the document to review"
+        },
+        "email_context": {
+          "type": "object",
+          "description": "For email mode: { recipient_type: 'client'|'opposing_counsel'|'court'|'regulatory'|'third_party', subject, prior_thread, urgency: 'routine'|'urgent'|'time_sensitive', tone: 'formal'|'firm'|'collaborative'|'aggressive' }",
+          "properties": {
+            "recipient_type": { "type": "string", "enum": ["client", "opposing_counsel", "court", "regulatory", "third_party"] },
+            "subject": { "type": "string" },
+            "prior_thread": { "type": "string" },
+            "urgency": { "type": "string", "enum": ["routine", "urgent", "time_sensitive"] },
+            "tone": { "type": "string", "enum": ["formal", "firm", "collaborative", "aggressive"] }
+          }
+        },
+        "special_instructions": {
+          "type": "string",
+          "description": "Any additional instructions, constraints, or preferences for the output"
+        },
+        "include_citations": {
+          "type": "boolean",
+          "description": "Whether to include full Bluebook citations. Defaults to true for research and exam modes."
+        },
+        "output_format": {
+          "type": "string",
+          "enum": ["markdown", "plain_text", "html", "docx_template"],
+          "description": "Desired output format. Defaults to markdown."
+        }
+      }
+    },
+    "outputs_schema": {
+      "type": "object",
+      "properties": {
+        "work_product": {
+          "type": "string",
+          "description": "The primary legal work product (memo, document, email, analysis, etc.)"
+        },
+        "summary": {
+          "type": "string",
+          "description": "Executive summary of the work product and key conclusions"
+        },
+        "risk_assessment": {
+          "type": "object",
+          "description": "Risk analysis: { overall_risk: 'low'|'medium'|'high'|'critical', factors: [], mitigation: [] }",
+          "properties": {
+            "overall_risk": { "type": "string", "enum": ["low", "medium", "high", "critical"] },
+            "factors": { "type": "array", "items": { "type": "string" } },
+            "mitigation": { "type": "array", "items": { "type": "string" } }
+          }
+        },
+        "citations": {
+          "type": "array",
+          "description": "List of legal citations referenced in the work product",
+          "items": { "type": "string" }
+        },
+        "action_items": {
+          "type": "array",
+          "description": "Recommended next steps or follow-up actions",
+          "items": { "type": "string" }
+        },
+        "confidence_level": {
+          "type": "string",
+          "enum": ["high", "moderate", "low", "uncertain"],
+          "description": "Confidence level in the analysis or recommendation"
+        },
+        "disclaimers": {
+          "type": "array",
+          "description": "Applicable legal disclaimers and caveats",
+          "items": { "type": "string" }
+        }
+      }
+    },
+    "bindings": [
+      {
+        "type": "community",
+        "source": "lincoln-ai",
+        "category": "Productivity"
+      }
+    ],
+    "constraints": {
+      "riskTier": 3,
+      "importedStub": false
+    },
+    "required_permissions": [
+      "file_read",
+      "file_write",
+      "network_access"
+    ],
+    "examples": [
+      {
+        "input": "{ \"mode\": \"draft\", \"document_type\": \"NDA\", \"jurisdiction\": \"DE\", \"parties\": { \"client\": { \"name\": \"Acme Corp\", \"role\": \"Disclosing Party\" }, \"opposing\": { \"name\": \"Beta LLC\", \"role\": \"Receiving Party\" } }, \"special_instructions\": \"Mutual NDA, 2-year term, carve-out for independently developed information\" }",
+        "output": "MUTUAL NON-DISCLOSURE AGREEMENT\n\nThis Mutual Non-Disclosure Agreement (\"Agreement\") is entered into as of [DATE] by and between Acme Corp, a Delaware corporation (\"Acme\"), and Beta LLC, a [State] limited liability company (\"Beta\"). Each party may be referred to as a \"Disclosing Party\" or \"Receiving Party.\" ...[full NDA with standard provisions, carve-outs, remedies, governing law, and signature blocks]"
+      },
+      {
+        "input": "{ \"mode\": \"exam\", \"query\": \"A enters B's property without permission and takes B's car. While driving away, A runs a red light and hits C, a pedestrian. C suffers a broken leg. Discuss all potential claims and defenses.\", \"practice_area\": \"torts\" }",
+        "output": "TORTS ANALYSIS — MULTI-PARTY HYPOTHETICAL\n\nI. A's LIABILITY TO B\n\nA. Trespass to Land: A entered B's property without permission, satisfying the elements of (1) intentional entry (2) onto land possessed by another (3) without consent. A is liable for all damages proximately caused, including consequential damages. Restatement (Second) of Torts §158.\n\nB. Trespass to Chattels / Conversion: Taking B's car constitutes conversion — an intentional exercise of dominion or control over B's personal property that so seriously interferes with B's right of control that A must pay the full value. Restatement (Second) §222A. Since A took the car entirely, this rises above mere trespass to chattels (§217) to full conversion.\n\nII. A's LIABILITY TO C\n\nA. Negligence: (1) Duty — A owed a duty of reasonable care to all foreseeable plaintiffs, including pedestrians (Palsgraf — Cardozo majority). (2) Breach — Running a red light violates traffic laws, establishing negligence per se under most jurisdictions... [continues with full IRAC analysis of all claims and defenses]"
+      },
+      {
+        "input": "{ \"mode\": \"email\", \"email_context\": { \"recipient_type\": \"opposing_counsel\", \"subject\": \"Settlement Demand — Smith v. Jones, Case No. 24-cv-1234\", \"tone\": \"firm\", \"urgency\": \"time_sensitive\" }, \"facts\": \"Our client was rear-ended at a stoplight. $45,000 in medical bills, 3 months lost wages ($15,000), ongoing physical therapy. Liability is clear from the police report. Insurance company offered $30,000 which is inadequate.\", \"jurisdiction\": \"CA\" }",
+        "output": "Subject: Settlement Demand — Smith v. Jones, Case No. 24-cv-1234\n\nCounsel:\n\nI write on behalf of my client, Jane Smith, regarding the above-referenced matter. Your client's offer of $30,000 is wholly inadequate given the documented damages and clear liability.\n\nMs. Smith's damages are as follows:\n- Medical expenses to date: $45,000 (supported by itemized billing records)\n- Lost wages: $15,000 (3 months, verified by employer)\n- Future medical treatment: estimated $12,000 (per treating physician)\n- Pain and suffering: $60,000 (conservative multiplier given the duration and severity)\n\nTotal demand: $132,000\n\nLiability is undisputed. The police report confirms your client rear-ended Ms. Smith while she was stopped at a red light. California follows pure comparative negligence (Li v. Yellow Cab Co., 13 Cal.3d 804 (1975)), and there is zero comparative fault on our client's part.\n\nWe are prepared to file a motion for summary judgment on liability if this matter does not resolve within 21 days. Please respond with a substantive counteroffer by [DATE].\n\nVery truly yours,\n[Agent Name]"
+      },
+      {
+        "input": "{ \"mode\": \"research\", \"query\": \"Can a smart contract on a public blockchain constitute a 'security' under federal securities law? What is the current state of the law after SEC v. Ripple?\", \"practice_area\": \"crypto_web3\", \"jurisdiction\": \"federal\", \"include_citations\": true }",
+        "output": "LEGAL MEMORANDUM\n\nQUESTION PRESENTED: Whether a smart contract token on a public blockchain constitutes an 'investment contract' (and thus a 'security') under Section 2(a)(1) of the Securities Act of 1933.\n\nSHORT ANSWER: It depends on the totality of circumstances under the Howey test. Post-Ripple, the analysis turns on the manner of sale and reasonable expectations of purchasers, not the nature of the underlying technology...\n\n[Full CREAC memo with Howey analysis, SEC v. W.J. Howey Co., 328 U.S. 293 (1946), SEC v. Ripple Labs, No. 20-cv-10832 (S.D.N.Y. 2023), including programmatic sales vs. institutional sales distinction, SEC v. Terraform Labs, SEC v. Coinbase, and current regulatory landscape]"
+      }
+    ],
+    "eval_packs": [],
+    "provenance": {
+      "publisher": "apeclaw",
+      "signed": false,
+      "source": "custom_skill",
+      "sourceUrl": "https://github.com/simplefarmer69/ape-claw",
+      "homepage": "https://openclaw.ai",
+      "license": "MIT",
+      "importedAt": "2026-02-20T13:30:00.000Z"
+    },
+    "vettedOk": true,
+    "onchainTokenId": "10892",
+    "onchainMintTx": "0xc9e4b906065f422791c8330d02422dc6c56eccf071407002984d13a5840731c6",
+    "onchainPublishTx": "0x75b5bc2d9a5bf81260786896cac454e0b782f59caf863c3e1ed7e9aa4bcade56"
+  }
+}

package/docs/APECLAW_V2_ALPHA.md

@@ -192,7 +192,7 @@ This is the minimum substrate for crash recovery and persistent operation.
 If you don't have `ape-claw` on PATH yet, you can run v2 commands via:
 
 ```bash
-npx --yes github:simplefarmer69/ape-claw pod init --dir ./pod-workspace --json
+npx ape-claw pod init --dir ./pod-workspace --json
 ```
 
 ## Compatibility guardrails (do not break v1)

package/docs/CLI_GUIDE.md

@@ -9,7 +9,7 @@ The UI is optional; the CLI + onchain receipts are the backbone.
 Most docs use `npx` so you can run without a global install:
 
 ```bash
-npx --yes github:simplefarmer69/ape-claw doctor --json
+npx ape-claw doctor --json
 ```
 
 If you already have `ape-claw` installed globally, use it directly.