[{"_1":2,"_3":-5,"_4":-5},"loaderData",{"_5":6,"_7":8,"_9":10},"actionData","errors","root",{"_110":111},"routes/modules.$creator.$moduleId",{"_14":15,"_16":17,"_18":19,"_20":21},"routes/modules.$creator.$moduleId.endpoints",{"_11":12,"_13":-5},"error","Module not found","tools","module",{"_20":22,"_23":24,"_25":26,"_27":28,"_29":30,"_31":32,"_33":26,"_34":-5,"_35":36,"_37":-5,"_38":-5,"_39":-5,"_40":41,"_42":43,"_44":45,"_46":45,"_47":48,"_49":50,"_51":52,"_53":-5,"_54":45,"_55":56,"_57":58,"_59":-5,"_60":-5,"_61":-5,"_62":-5,"_63":64,"_65":66,"_67":68,"_69":70,"_71":72,"_73":-5,"_74":-5,"_75":-5,"_76":45,"_77":-5,"_78":79,"_80":81,"_82":83,"_84":85,"_86":87,"_88":89,"_90":91},"apiUrl","https://odel.app/api","creator","johnny-z13","moduleId","proofslip","@johnny-z13/proofslip","displayName","ProofSlip","tagline","Receipt-based verification for AI agent workflows — create, verify, and poll ephemeral proof objects","iconKey","https://avatars.githubusercontent.com/u/184745011?v=4","category","ai-agents","status","published","metaDescription","metaKeywords","deploymentType","external","workerName","bindingName","mcpConfig","pricingModel","free","creatorUserId","system:external-import","createdAt",1783311704000,"updatedAt","sourceRegistry","official-mcp-registry","sourceUrl","https://registry.modelcontextprotocol.io/v0/servers?search=ai.proofslip%2Fmcp-server","sourceVerified",false,"sourceMissingSince","importedAt","upstreamStars",1,"upstreamUpdatedAt",1775572186000,"detectedTransport","detectedAuthType","probeStatus","lastProbedAt","githubRepoUrl","https://github.com/Johnny-Z13/proofslip","githubRepoFullName","Johnny-Z13/proofslip","releaseId","ext-johnny-z13-proofslip","version","0.2.3","descriptionMarkdown","# ProofSlip\n\n**Portable proof objects for agent workflows.**\n\n24-hour ephemeral receipts that let agents verify what happened before deciding what happens next. Create a receipt, check it later, let it expire. That's it.\n\n> Agents should not continue based on assumptions when they can continue based on receipts.\n\n[Live Site](https://proofslip.ai) | [API Docs](https://proofslip.ai/docs) | [Example Receipt](https://proofslip.ai/example) | [MCP Server](https://www.npmjs.com/package/@proofslip/mcp-server) | [llms.txt](https://proofslip.ai/llms.txt)\n\n## The Problem\n\nAgentic workflows break in predictable ways: duplicate side effects, unclear approval states, unsafe retries, ambiguous resumability. Teams currently solve this with raw logs, brittle flags, ad hoc DB rows, and hand-rolled retry logic.\n\nProofSlip replaces all of that with a single primitive: a short-lived, machine-readable receipt — a **portable proof object** that travels with the workflow and answers the questions agents actually ask:\n\n- Did this action already happen?\n- Was the request approved?\n- Is retry safe?\n- Should I continue, wait, or escalate?\n\n## How It Works\n\n```\n1. Agent completes a step → creates a receipt (POST /v1/receipts)\n2. Next agent (or same agent later) → verifies the receipt before acting\n3. Receipt expires after 24 hours → no cleanup burden\n```\n\nEvery receipt includes polling guidance (`is_terminal`, `next_poll_after_seconds`) so agents know whether to stop, wait, or keep checking.\n\n## Receipt Types\n\n| Type | Use Case |\n|------|----------|\n| `action` | Record that something happened (always terminal) |\n| `approval` | Track pending/approved/rejected decisions |\n| `handshake` | Two agents confirming shared context |\n| `resume` | Bookmark for safe workflow continuation |\n| `failure` | Structured failure with retry guidance (always terminal) |\n\n## Quick Start\n\n### 1. Get an API Key\n\n```bash\ncurl -X POST https://proofslip.ai/v1/auth/signup \\\n -H \"Content-Type: application/json\" \\\n -d '{\"email\": \"you@example.com\"}'\n```\n\nSave the returned key immediately — it cannot be retrieved later.\n\n### 2. Create a Receipt\n\n```bash\ncurl -X POST https://proofslip.ai/v1/receipts \\\n -H \"Authorization: Bearer ak_your_key_here\" \\\n -H \"Content-Type: application/json\" \\\n -d '{\n \"type\": \"action\",\n \"status\": \"success\",\n \"summary\": \"Sent welcome email to user@example.com\",\n \"ref\": {\n \"workflow_id\": \"onboarding-123\",\n \"agent_id\": \"email-agent\"\n }\n }'\n```\n\nReturns:\n\n```json\n{\n \"receipt_id\": \"rct_abc123...\",\n \"type\": \"action\",\n \"status\": \"success\",\n \"summary\": \"Sent welcome email to user@example.com\",\n \"verify_url\": \"https://proofslip.ai/verify/rct_abc123...\",\n \"created_at\": \"2026-03-25T12:00:00.000Z\",\n \"expires_at\": \"2026-03-26T12:00:00.000Z\",\n \"is_terminal\": true,\n \"next_poll_after_seconds\": null\n}\n```\n\n### 3. Verify Before Acting\n\n```bash\n# JSON (for agents)\ncurl https://proofslip.ai/v1/verify/rct_abc123?format=json\n\n# HTML (for humans — paste the verify_url in a browser)\n```\n\n### 4. Poll Non-Terminal Receipts\n\nFor approval workflows where a receipt starts as `pending`:\n\n```bash\n# Lightweight status check (no payload, no ref — just state)\ncurl https://proofslip.ai/v1/receipts/rct_abc123/status\n```\n\n```json\n{\n \"receipt_id\": \"rct_abc123...\",\n \"status\": \"pending\",\n \"is_terminal\": false,\n \"next_poll_after_seconds\": 30,\n \"expires_at\": \"2026-03-26T12:00:00.000Z\"\n}\n```\n\n## API Reference\n\n| Method | Endpoint | Auth | Description |\n|--------|----------|------|-------------|\n| `POST` | `/v1/receipts` | API Key | Create a receipt |\n| `GET` | `/v1/verify/:id` | None | Verify a receipt (JSON or HTML) |\n| `GET` | `/verify/:id` | None | Shortcut verify (same behavior) |\n| `GET` | `/v1/receipts/:id/status` | None | Lightweight polling endpoint |\n| `POST` | `/v1/auth/signup` | None | Get an API key |\n| `GET` | `/health` | None | Health check |\n\n### Create Receipt Fields\n\n| Field | Required | Description |\n|-------|----------|-------------|\n| `type` | Yes | `action`, `approval`, `handshake`, `resume`, `failure` |\n| `status` | Yes | Any string (e.g. `success`, `pending`, `rejected`) |\n| `summary` | Yes | Human-readable description, max 280 chars |\n| `payload` | No | Structured JSON data, max 4KB |\n| `ref` | No | Workflow references: `run_id`, `agent_id`, `action_id`, `workflow_id`, `session_id` |\n| `expires_in` | No | TTL in seconds (60–86400). Default: 86400 (24h) |\n| `idempotency_key` | No | Prevents duplicate receipts on retry |\n| `audience` | No | Set to `\"human\"` to enrich the verify page with OG/Twitter social card meta tags |\n\n### Rate Limits\n\n| Endpoint | Limit | Scope |\n|----------|-------|-------|\n| `POST /v1/receipts` | 60/min | Per API key |\n| `GET /v1/verify/:id` | 120/min | Per IP |\n| `GET /v1/receipts/:id/status` | 120/min | Per IP |\n| `POST /v1/auth/signup` | 5/min | Per IP |\n\nRate limit headers (`X-RateLimit-Limit`, `X-RateLimit-Remaining`, `X-RateLimit-Reset`) are included on every response.\n\n### Idempotency\n\nInclude an `idempotency_key` to safely retry receipt creation. If a receipt with the same key already exists and the content matches, the original receipt is returned. If the content differs, you'll get a `409 idempotency_conflict` error.\n\n## MCP Server\n\nUse ProofSlip as a tool in Claude Desktop, Cursor, Windsurf, or any MCP-compatible client:\n\n```bash\nnpx -y @proofslip/mcp-server\n```\n\n**Tools available:**\n\n| Tool | Description |\n|------|-------------|\n| `create_receipt` | Create a verifiable receipt |\n| `verify_receipt` | Verify a receipt by ID |\n| `check_status` | Lightweight status poll |\n\n**Configuration (Claude Desktop `claude_desktop_config.json`):**\n\n```json\n{\n \"mcpServers\": {\n \"proofslip\": {\n \"command\": \"npx\",\n \"args\": [\"-y\", \"@proofslip/mcp-server\"],\n \"env\": {\n \"PROOFSLIP_API_KEY\": \"ak_your_key_here\"\n }\n }\n }\n}\n```\n\n**Cursor / Windsurf:** Add the same config to your MCP settings file.\n\n## Agent Discovery\n\nProofSlip exposes machine-readable discovery endpoints so agents and tools can self-onboard:\n\n| Endpoint | Format | Purpose |\n|----------|--------|---------|\n| [`/llms.txt`](https://proofslip.ai/llms.txt) | Plain text | LLM context summary |\n| [`/llms-full.txt`](https://proofslip.ai/llms-full.txt) | Plain text | Complete API reference for LLMs |\n| [`/docs`](https://proofslip.ai/docs) | HTML | Human-readable API docs |\n| [`/.well-known/openapi.json`](https://proofslip.ai/.well-known/openapi.json) | JSON | OpenAPI 3.1 spec |\n| [`/.well-known/ai-plugin.json`](https://proofslip.ai/.well-known/ai-plugin.json) | JSON | ChatGPT plugin manifest |\n| [`/.well-known/mcp.json`](https://proofslip.ai/.well-known/mcp.json) | JSON | MCP server discovery |\n| [`/.well-known/agent.json`](https://proofslip.ai/.well-known/agent.json) | JSON | Agent protocol discovery |\n\n## Self-Hosting\n\n### Prerequisites\n\n- Node.js 18+\n- PostgreSQL (Neon recommended, any Postgres works)\n\n### Setup\n\n```bash\ngit clone https://github.com/Johnny-Z13/proofslip.git\ncd proofslip\nnpm install\n\n# Configure environment\ncp .env.example .env\n# Edit .env with your DATABASE_URL and BASE_URL\n\n# Run migrations\nnpm run db:migrate\n\n# Seed an API key\nnpm run db:seed -- you@example.com\n\n# Start dev server\nnpm run dev\n```\n\n### Environment Variables\n\n| Variable | Required | Description |\n|----------|----------|-------------|\n| `DATABASE_URL` | Yes | Postgres connection string |\n| `BASE_URL` | Yes | Public URL (used in `verify_url` generation) |\n| `CRON_SECRET` | Production | Bearer token for the cleanup cron endpoint |\n| `RESEND_API_KEY` | Production | Resend API key for emailing API keys to web signups |\n| `DEV_SECRET` | Optional | Secret key to access `/dev/console` test page |\n| `NODEJS_HELPERS` | Production | Set to `0` (required for Hono zero-config on Vercel) |\n\n### Scripts\n\n| Command | Description |\n|---------|-------------|\n| `npm run dev` | Start local dev server (port 3000, hot reload) |\n| `npm test` | Run test suite |\n| `npm run db:generate` | Generate migrations from schema changes |\n| `npm run db:migrate` | Run database migrations |\n| `npm run db:seed -- ` | Create an API key for the given email |\n\n## Deployment Architecture\n\n```mermaid\ngraph TB\n subgraph \"Your Machine\"\n GIT[\"Git Repo
(source of truth)\"]\n end\n\n subgraph \"GitHub\"\n REPO[\"github.com/Johnny-Z13/proofslip\"]\n end\n\n subgraph \"Vercel\"\n AUTO[\"Auto-Build on Push\"]\n API[\"Serverless Functions
(Hono API)\"]\n LP[\"Landing Page
(served from API)\"]\n CRON[\"Hourly Cron
POST /cron/cleanup\"]\n end\n\n subgraph \"Neon\"\n DB[(\"PostgreSQL
receipts + api_keys\")]\n end\n\n GIT -- \"git push\" --> REPO\n REPO -- \"webhook triggers\" --> AUTO\n AUTO --> API\n AUTO --> LP\n API -- \"DATABASE_URL\" --> DB\n CRON -- \"deletes expired\" --> DB\n\n style GIT fill:#1a1a2e,color:#fff\n style REPO fill:#161b22,color:#fff\n style AUTO fill:#000,color:#fff\n style API fill:#000,color:#fff\n style LP fill:#000,color:#fff\n style CRON fill:#000,color:#fff\n style DB fill:#0a2e1a,color:#fff\n```\n\n**Environment variables on Vercel:**\n| Variable | Where to set |\n|----------|-------------|\n| `DATABASE_URL` | Vercel → Settings → Environment Variables (copy from Neon dashboard) |\n| `BASE_URL` | `https://proofslip.ai` (or your custom domain) |\n| `CRON_SECRET` | Generate a random token, set in Vercel env vars |\n| `NODEJS_HELPERS` | Set to `0` (required for Hono zero-config deployment) |\n\n## API Request/Response Flows\n\n```mermaid\nsequenceDiagram\n participant Agent as Agent / Client\n participant API as Vercel API\n participant DB as Neon Postgres\n\n note over Agent,DB: 1. Signup\n Agent->>API: POST /v1/auth/signup
{\"email\": \"...\"}\n API->>DB: INSERT api_keys\n API-->>Agent: {\"api_key\": \"ak_...\"}\n\n note over Agent,DB: 2. Create Receipt\n Agent->>API: POST /v1/receipts
Bearer ak_...
{\"type\",\"status\",\"summary\",...}\n API->>API: validate, rate-limit, auth\n API->>DB: INSERT receipts\n API-->>Agent: {\"receipt_id\": \"rct_...\"
\"verify_url\", \"is_terminal\",
\"next_poll_after_seconds\"}\n\n note over Agent,DB: 3a. Verify (full data)\n Agent->>API: GET /v1/verify/rct_...?format=json\n API->>DB: SELECT receipt\n API-->>Agent: {full receipt data}\n\n note over Agent,DB: 3b. Poll (lightweight)\n loop until is_terminal = true\n Agent->>API: GET /v1/receipts/rct_.../status\n API->>DB: SELECT status fields only\n API-->>Agent: {\"status\",\"is_terminal\",
\"next_poll_after_seconds\"}\n Note right of Agent: wait next_poll_after_seconds\n end\n\n note over API,DB: 4. Cleanup (hourly cron)\n API->>DB: DELETE WHERE expires_at < now()\n```\n\n## Architecture\n\n```\nsrc/\n├── index.ts # Hono app, middleware chain, routing\n├── dev.ts # Local dev server\n├── db/\n│ ├── schema.ts # Drizzle schema (receipts, api_keys)\n│ ├── client.ts # Neon DB connection\n│ └── migrate.ts # Migration runner\n├── routes/\n│ ├── receipts.ts # POST /v1/receipts\n│ ├── verify.ts # GET /verify/:id, /v1/verify/:id\n│ ├── status.ts # GET /v1/receipts/:id/status\n│ ├── auth.ts # POST /v1/auth/signup\n│ └── cron.ts # POST /cron/cleanup\n├── middleware/\n│ ├── api-key-auth.ts # Bearer token → hash validation\n│ ├── rate-limit.ts # Per-key and per-IP rate limiting\n│ ├── security.ts # CORS, headers, body limits\n│ └── logger.ts # Structured JSON request logging\n├── lib/\n│ ├── ids.ts # nanoid generators (rct_, key_, req_)\n│ ├── hash.ts # SHA-256 hashing\n│ ├── validate.ts # Input validation\n│ ├── polling.ts # Terminal state detection, poll intervals\n│ ├── rate-limit.ts # In-memory sliding window tracker\n│ └── errors.ts # Error response builder\n└── views/\n ├── font.ts # Departure Mono as inline base64\n ├── og-image.ts # Branded SVG for social cards\n ├── landing-page.ts # HTML homepage\n ├── docs-page.ts # Single-page API docs\n ├── verify-page.ts # Receipt display (shareable, OG tags when audience=human)\n ├── not-found-page.ts # 404 receipt page\n ├── llms-txt.ts # /llms.txt content\n ├── llms-full-txt.ts # /llms-full.txt complete reference\n ├── openapi.ts # OpenAPI 3.1 spec\n └── mcp-json.ts # MCP discovery manifest\n```\n\n**Stack**: Hono + Drizzle ORM + Neon Postgres + Vercel Serverless (zero-config)\n\n## Better With Context Capsule\n\nProofSlip proves what happened. [**Context Capsule**](https://contextcapsule.ai) tells agents what to do next. Together they're two primitives for reliable agent workflows:\n\n| | ProofSlip | Context Capsule |\n|---|---|---|\n| **Role** | Evidential | Navigational |\n| **Answers** | \"What actually happened, and can you verify it?\" | \"What's the situation and what should happen next?\" |\n| **Primitive** | Verifiable receipt | Execution context packet |\n| **Get started** | [proofslip.ai](https://proofslip.ai) | [contextcapsule.ai](https://contextcapsule.ai) |\n\n### Example: Verified Handoff\n\n```bash\n# 1. Agent A completes work and creates a receipt as proof\ncurl -X POST https://proofslip.ai/v1/receipts \\\n -H \"Authorization: Bearer ak_your_key\" \\\n -H \"Content-Type: application/json\" \\\n -d '{\n \"type\": \"action\",\n \"status\": \"success\",\n \"summary\": \"Migrated user table schema\"\n }'\n# Returns: { \"receipt_id\": \"rct_abc123\", ... }\n\n# 2. Agent A creates a Context Capsule referencing that receipt\ncurl -X POST https://contextcapsule.ai/v1/capsules \\\n -H \"Authorization: Bearer ck_your_key\" \\\n -H \"Content-Type: application/json\" \\\n -d '{\n \"summary\": \"Schema migration complete, ready for integration tests\",\n \"decisions\": [\"Used addColumn to preserve backward compat\"],\n \"next_steps\": [\"Run integration tests\", \"Update API serializers\"],\n \"refs\": {\n \"receipt_ids\": [\"rct_abc123\"],\n \"workflow_id\": \"migration-456\"\n }\n }'\n\n# 3. Agent B picks up the capsule, verifies the receipt, then continues\ncurl https://contextcapsule.ai/v1/capsules/cap_xyz789?format=json\ncurl https://proofslip.ai/v1/verify/rct_abc123?format=json\n# Receipt checks out → proceed with integration tests\n```\n\nThe receipt carries the proof. The capsule carries the navigation. Neither agent has to trust the other's claims — they verify.\n\n> **Don't have Context Capsule yet?** [Get a free API key](https://contextcapsule.ai/v1/auth/signup) — same stack, same patterns, takes 30 seconds.\n\n## Design Principles\n\n- **Ephemeral by default.** 24-hour TTL. Not a long-term archive.\n- **Machine + human readable.** JSON for agents, styled HTML for browsers. Same URL.\n- **Cheap to run.** Serverless, auto-expiring data, minimal storage footprint.\n- **Simple mental model.** Create. Verify. Expire. Three operations.\n- **Idempotent by design.** Safe retries built into the protocol.\n\n## License\n\n[MIT](https://github.com/Johnny-Z13/proofslip/blob/master/LICENSE)\n","changelogMarkdown","videoKey","requiredSecretsJson","publishedAt","priceCents","trialDays",0,"categoryDisplayName","AI Agents","upstream",{"_94":95,"_96":97,"_98":99,"_100":101,"_102":103},"requiredSecrets",[],"tags",[],"assets",[],"stats",{"_92":79,"_93":79},"ratingAvg","reviewCount","topics",[104,105,106,30,107,108,109],"license","MIT","language","TypeScript","registryName","ai.proofslip/mcp-server","homepage","https://proofslip.ai","agentic-workflow","agents","ai","mcp","reciept","reciept-generator","ENV",{"_112":113,"_114":115,"_116":117},"ENVIRONMENT","production","TURNSTILE_SITE_KEY","0x4AAAAAAB8-rEI2gNDf_QhU","SENTRY_DSN","https://45fd1229f78089d5c3ee78be128030e1@o4510583261757440.ingest.de.sentry.io/4510584253448272"]