{
  "markdown": "# TradeOS Public Intelligence Kit\n\nBuild source-grounded crypto market Data Intelligence products on top of\n[TradeOS public evidence](https://tradeos.tech/market).\n\nTradeOS is the crypto-market vertical of a broader source-grounded Data\nIntelligence OS strategy. It is not a trading platform, broker, custodian,\nmanaged account service, or hosted execution system. See the\n[TradeOS Business Thesis](docs/tradeos-business-thesis.md) for the business\noverview.\n\nThis repo gives builders SDKs, MCP tools, a BYOK agent CLI, runnable bots, and\nthe blueprint for the flagship TradeOS app pattern: a **private self-hosted\nSymbol Cockpit**.\n\nTradeOS hosts the intelligence API. The cockpit is an individual private-use\ncontrol plane: it runs on the self-hosted operator's own machine, server, quant\nworkstation, or agent host. It can recommend buy, sell, trim, avoid, watch, or\npass decisions from TradeOS evidence plus local rules, while keys, custody,\nportfolio context, approvals, and execution stay in that operator's environment.\n\nBuilders can package or extend this pattern for users to self-host. They should\nnot operate the cockpit as a managed account service, collect customer exchange\ncredentials, custody customer assets, or place orders for customers.\n\nThat private self-hosted model is the product advantage. Traders get an\nactionable agent. Builders get a product they can ship. TradeOS stays the\nsource-grounded intelligence product layer instead of becoming a custodian,\nbroker, or hosted order router.\n\nThink of TradeOS as a crypto market Data Intelligence OS:\n\n- TradeOS supplies the source-backed intelligence layer.\n- Builders package that intelligence into briefings, bots, watchlists,\n  dashboards, validation packs, proof pages, and widgets.\n- Users and builder apps send structured feedback that improves intelligence\n  quality and provenance.\n- Paid TradeOS, x402, or enterprise access starts when a product needs scale,\n  alerts, automation, premium history, private intelligence products,\n  validation APIs, or explicit data rights.\n\nThe practical loop is simple:\n\n```text\nUse TradeOS free.\nEarn Data Intel Credits by improving intelligence quality.\nBuild and earn on public intelligence.\nPay when you need private intelligence products, scale, alerts, automation, or data rights.\n```\n\nUse the free public kit to prove a workflow. Charge for the packaging around\nthat workflow. Submit feedback with provenance so TradeOS can improve the\nintelligence layer. Upgrade to paid TradeOS when the product needs production\nvolume, alerts, automation, premium history, private intelligence products,\nvalidation APIs, x402 machine access, explicit data rights, or enterprise\nsupport.\n\nThe flywheel effect is the point: useful TradeOS evidence helps builders ship\npaid services; paid services create users, usage, feedback, and outcome labels;\nthat feedback improves TradeOS intelligence; better intelligence makes builder\nproducts more valuable; successful products then need paid TradeOS depth, x402\naccess, and enterprise support.\n\n## Data Intelligence Services / SKU Map\n\nTradeOS packages intelligence as product layers, not only as raw datasets:\n\n| Product Layer | Builder-Facing Shape | Access Model |\n| --- | --- | --- |\n| Public intelligence products | Market Pulse, Platform Pulse, Token Radar, Fusion Signal Lite, Ask TradeOS, watchlists, and dossiers | Free public baseline with bounded read-only context |\n| DTI credit unlock SKUs | Refresh boost, symbol pack, history window, evidence depth, Token Discovery detail, Fusion Signal detail, Review Lab passes, and AskTradeOS question packs | Human Data Intel Credits for temporary public GUI depth, public Ask packs, or read-only Review Lab access |\n| Private intelligence passes | `private_30m`, `private_1h`, and `private_4h` private dashboard windows | Exact x402 pay-per-view sessions for private intelligence surfaces |\n| Agent/API/x402 data SKUs | Agent router, Venice raw inference, Venice intel router, token risk, token discovery, risk-gated discovery, signal quality, signal evidence packs, fusion snapshots/history, Market Pulse Pro/team, VPIN stress, dataset concierge, and package scoping | Paid or reviewed production access for builders, agents, marketplaces, and API consumers |\n| NEAR Intents-adjacent x402 wrappers | risk preflight, route context, asset dossier, status monitor, liquidity caveats, solver readiness, and dataset scope | Same read-only SKU family packaged for intent-aware apps; x402 pays for intelligence only, while NEAR Intents or the partner app handles execution |\n\nKeep the boundary explicit in builder products. Human DTI credits do not unlock\npaid API scale, x402 calls, alert delivery, exports, automation, custody,\nexecution, private context, or enterprise data rights. Those are paid, reviewed,\nor contract-entitled paths.\n\n## At A Glance\n\n| Topic | Default |\n| --- | --- |\n| Public API | `https://api.tradeos.tech/v1/public-intel` |\n| Business thesis | TradeOS is the crypto-market Data Intelligence OS vertical; see [TradeOS Business Thesis](docs/tradeos-business-thesis.md) |\n| Public reads | no TradeOS account required, server-side limits apply |\n| Saved watchlists | require a signed-in TradeOS account token |\n| Builder attribution | optional `TRADEOS_PUBLIC_INTEL_KEY` |\n| DTI/API lifecycle | [Data Intel Credit Loop](docs/feedback-credit-loop.md) explains human DTI, app reputation DTI, and feedback status endpoints |\n| LLM inference | recommended BYOK default: [Venice AI](https://venice.ai/pricing) with `e2ee-glm-5-1` for privacy-enhanced self-hosted inference; TradeOS also publishes paid AntSeed/x402 Venice raw-inference and evidence-backed intel-router SKUs |\n| Paid machine access | x402 payment or TradeOS entitlement |\n| Live platform proof | [Platform Pulse](https://tradeos.tech/market) shows feedback signals, x402 challenge demand, source mix, and settlement health |\n| Public Fusion preview | `https://tradeos.tech/fusion` shows bounded symbol-level Signal Cockpit reads and feedback intake; see [Public Fusion Signal Cockpit Preview](docs/public-fusion-signal-cockpit-preview.md) |\n| Earn as a builder | package TradeOS intelligence into paid services, workflows, agents, or vertical apps; start with the [Earn as a Builder Field Guides](docs/commercial-distribution/README.md) |\n| Flagship model | individual private-use, self-hosted Symbol Cockpit over TradeOS intelligence |\n| Cockpit trading-intelligence universe | 21 full cockpit symbols today, plus 52 public thesis registry symbols; see [Symbol Intelligence Coverage](docs/symbol-intelligence-coverage.md) |\n| Safety boundary | actionable recommendations are allowed; TradeOS does not custody keys, manage accounts, place orders, or support third-party account management |\n\n## What This Is\n\nTradeOS sits in the public market-intelligence layer for crypto and on-chain\nmarkets. It is one vertical, not the entire Source Intelligence Network\nstrategy. This kit is the builder distribution surface for that Data\nIntelligence OS. It helps products discover, consume, package, and send\nfeedback on current, source-backed market context inside bots, dashboards,\nagent workflows, research products, validation systems, and paid communities.\n\nIt is an evidence and feedback layer:\n\n- public digest inputs, candidates, watchlists, thesis records, proof lookups,\n  caveats, freshness, source refs, and invalidation notes;\n- a public Fusion Signal Cockpit preview that shows redacted symbol-level\n  direction, confidence, freshness, why-fired context, invalidation, and\n  feedback intake without exposing execution fields;\n- symbol-cockpit and action-agent patterns for turning evidence into\n  good/bad/ugly verdicts, watchlist recommendations, and bot preflight checks;\n- non-executable action intents that carry evidence-backed action context into\n  local review, policy gates, paper execution, or independent executor\n  experiments without becoming orders;\n- TypeScript and Python SDKs for apps and services;\n- a stdio MCP server for local agent hosts such as Claude Desktop and Cursor;\n- a BYOK CLI that can ask Venice AI or another OpenAI-compatible model over\n  TradeOS evidence;\n- a market briefing bot that can post to stdout, Discord, or Telegram;\n- structured feedback writes with provenance so TradeOS can learn which public\n  intelligence was useful, early, late, thin, or confusing.\n\nIt is not a TradeOS-hosted broker, custody product, managed account service, or\norder router. Builders can wire the open-source kit into local bot, preflight,\nor execution stacks, but exchange keys, approvals, sizing rules, and custody\nstay with the self-hosted operator using it for their own account context. If a\nbuilder collects customer credentials, controls customer accounts, or runs\nexecution for other people, that builder has created a separate managed trading\nservice outside this kit's supported boundary.\n\nAction intents are the bridge, not the executor. They are marked\n`non_executable`, require operator review, and deliberately omit venue, account,\nsize, order type, route, calldata, transaction body, and execute URL fields.\n\nIn practical terms:\n\n```text\nTradeOS Data Intelligence OS -> intelligence, evidence, data, feedback IDs\nSymbol Cockpit -> private local recommendation and decision runtime\nLocal modules -> feasibility, EA/risk, execution adapters, ops dashboard\n```\n\n## Why Builders Use It\n\n- **Avoid weak intelligence gaps**: reduce wrong-token routing, stale context,\n  unsupported claims, caveat-free recommendations, missed watchlist changes,\n  and outputs that cannot be audited. See [Problem Space](docs/problem-space.md)\n  for exposure anchors that translate weak context into potential dollar impact\n  without promising guaranteed loss prevention.\n- **Understand the business thesis**: TradeOS is the crypto-market Data\n  Intelligence OS vertical. It creates demand with useful public intelligence\n  and monetizes paid depth when workflows need scale, private intelligence,\n  alerts, validation, automation-safe access, or data rights. See\n  [TradeOS Business Thesis](docs/tradeos-business-thesis.md).\n- **Ship faster**: start with bounded TradeOS evidence instead of building a\n  market data pipeline first.\n- **Ground agents**: give LLMs current evidence, caveats, stable IDs, and source\n  references before they answer.\n- **Use privacy-enhanced BYOK inference**: use [Venice AI](https://venice.ai/pricing)\n  as the recommended default with `e2ee-glm-5-1`, an E2EE/TEE-backed upstream\n  model option. Builders can still swap in another OpenAI-compatible provider.\n  Builders that want a marketplace-hosted path can use paid TradeOS\n  AntSeed/x402 SKUs:\n  `tradeos-venice-raw-inference` for ungrounded model output, or\n  `tradeos-venice-intel-router` for TradeOS evidence-backed model summaries.\n- **Earn on top of TradeOS**: package TradeOS intelligence into services,\n  workflows, agents, dashboards, vertical apps, or research products customers\n  already understand and will pay to use.\n- **Show market proof**: point prospects to\n  [Platform Pulse](https://tradeos.tech/market) for live feedback volume, x402\n  challenge demand, source attribution, and settlement health. Treat x402\n  challenges as interest signals; only verified/completed payments are revenue.\n- **Distribute commercially**: use the commercial field guides to package\n  TradeOS-backed intelligence on Virtuals ACP, AntSeed, x402 directories, and\n  Agentic.Market-style discovery, plus NEAR Intents-adjacent preflight and\n  route-context surfaces, without publishing private keys or live provider\n  secrets.\n- **Sell privacy as the feature**: ship a cockpit that customers can run on\n  their own box, with their own model key, exchange keys, approval policy,\n  portfolio context, and audit logs.\n- **Close the loop**: send human, agent, or automation feedback back to TradeOS\n  with provenance.\n- **Scale cleanly**: graduate from public reads to paid TradeOS data, x402, or\n  enterprise access when the workflow needs scale, alerts, automation, private\n  intelligence products, or data rights.\n\n## What You Can Build\n\n| Product | Who Pays | What You Package | TradeOS Upgrade Trigger |\n| --- | --- | --- | --- |\n| Private symbol cockpit | active traders, research desks, token teams | good/bad/ugly verdicts, recommendation inbox, watchlist scanners | alerts, private intelligence context, larger universe, premium history |\n| Paid research digest | traders, funds, paid communities | edited briefings, archive, member access | more volume, custom universe, premium pulse |\n| Discord or Telegram market bot | token communities, DAOs, trading groups | server commands, scheduled summaries, feedback loop | high-volume reads, hosted bridge, team access |\n| Watchlist monitor | researchers, token teams, active traders | saved lists, risk changes, freshness, alerts, dashboard seats | alert delivery, webhooks, custom universes |\n| Tradebot intelligence layer | bot builders, quant developers | regime context, caveats, invalidation notes, post-trade review | automation-safe API, private forecasts, premium context |\n| Quant validation pack | quant shops, funds, signal vendors | outside labels, disagreement reports, replay/evaluation workflow | historical data, validation API, enterprise contract |\n| Claim or proof explorer | analysts, creators, research desks | public profile, proof pages, outcome tracking | paid proof tooling, creator analytics |\n| Market context widget | wallets, dashboards, data portals | risk caveats, digest context, source-backed snippets | B2B widget or API subscription |\n| Feedback analytics | research communities, analyst teams | label collection, reviewer workflow, app reputation | team analytics, deeper evidence, paid exports |\n\nThe public kit is the integration and discovery surface. Paid TradeOS starts\nwhen customers ask for production-grade volume, automation, history, delivery,\nprivate intelligence products, or explicit paid entitlement.\n\nMore product detail:\n\n- [Problem Space](docs/problem-space.md)\n- [Repository Layout](docs/repository-layout.md)\n- [Integration Keys And URLs](docs/integration-keys-and-urls.md)\n- [Getting API Keys And Requesting Scale](docs/getting-api-keys-and-scale.md)\n- [Flagship Symbol Cockpit](docs/flagship-symbol-cockpit.md)\n- [Symbol Cockpit And Action Agent](docs/symbol-cockpit-agent.md)\n- [Symbol Intelligence Coverage](docs/symbol-intelligence-coverage.md)\n- [Data Intelligence Product Model](docs/marketplace-model.md)\n- [Use Cases](docs/use-cases.md)\n- [Monetization Guide](docs/monetization.md)\n- [Earn as a Builder: Commercial Distribution Field Guides](docs/commercial-distribution/README.md)\n- [Builder Revenue Playbook](docs/builder-revenue-playbook.md)\n\n## Flagship: Symbol Cockpit\n\nFor a consumer-facing walkthrough of what the cockpit is, how to read it,\nrequirements, dependencies, launch steps, and safety boundaries, see\n[Flagship Symbol Cockpit](docs/flagship-symbol-cockpit.md).\n\nThe fastest consumer story is:\n\n```text\nGive TradeOS a symbol.\nTradeOS returns the good, bad, ugly, verdict, evidence, and an action\nrecommendation.\n```\n\nExample cockpit language:\n\n```text\nVVV: avoid new long.\nGood: momentum improved and sector interest remains present.\nBad: fusion agreement degraded and liquidity depth is thin.\nUgly: flow stress is elevated during broader market risk.\nRecommendation: avoid a fresh long; if already exposed, consider trim or tighter\nrisk controls until flow stress normalizes and fusion recovers.\nFeedback: useful / wrong / late / missing context.\n```\n\nThis cockpit can run privately in the operator's self-hosted deployment. Local\nwatchlists, strategy notes, wallet context, bot rules, and logs stay local\nunless the operator chooses to send feedback or authenticated context to\nTradeOS. TradeOS sees the public-intelligence queries and paid scopes the\nruntime sends.\n\nThat is why the cockpit is useful: it is not only a market summary. It is a\nself-hosted decision layer that can recommend buy, sell, trim, avoid, watch, or\npass based on TradeOS evidence plus the operator's local rules. The self-hosted\noperator owns what happens next.\n\nThis is the flagship architecture TradeOS wants builders to copy:\n\n```text\nTradeOS public or paid intelligence\n        |\n        v\nPrivate self-hosted cockpit\n        |\n        v\nNon-executable action intent -> local feasibility gate -> local EA/risk gate\n        |\n        v\nOptional local execution adapter\n        |\n        v\nUser-owned wallet, exchange, or broker account\n```\n\nFuture open-source modules can add feasibility checks, expected-advantage\nchecks, execution adapters, and a light operations dashboard. The control plane\nstill belongs to the self-hosted operator, not TradeOS.\n\n## Five-Minute Paths\n\nClone and build from source:\n\n```bash\ngit clone git@github.com:agenticsrclab/tradeos-public-intel-kit.git\ncd tradeos-public-intel-kit\nnpm install\nnpm run build\n```\n\n### Run The Symbol Cockpit\n\nThe flagship self-hosted app now lives in `apps/symbol-cockpit` and exposes the\nweb/API/worker runtime described in the cockpit ADR.\n\n```bash\nexport TRADEOS_PUBLIC_INTEL_KEY=<optional-public-intel-app-key>\nnpm run symbol-cockpit\n```\n\nOpen `http://127.0.0.1:18100`, or run the app-level Compose stack:\n\n```bash\ncd apps/symbol-cockpit\ncp .env.example .env\ndocker compose up\ndocker compose --profile risk up\ndocker compose --profile execution up\n```\n\nCLI and MCP consumers can use the same cockpit contracts:\n\n```bash\nnpm run cli -- cockpit VVV --chain 8453 --mode trader\nnpm run cli -- preflight VVV --action buy --chain 8453\n```\n\n### Run The Market Briefing Bot\n\nThis is the first recommended bot to fork: a source-backed market briefing\nworker for stdout, Discord, or Telegram.\n\n```bash\nnpm run briefing-bot -- brief\n```\n\nThat works without a TradeOS account and without an LLM key. It prints a\ndeterministic briefing from live public evidence.\n\nUse Venice AI as the recommended privacy-enhanced BYOK path for a stronger\nnatural-language brief. The kit defaults to `e2ee-glm-5-1`. Get a key from the\n[Venice AI subscription page](https://venice.ai/pricing):\n\n```bash\nexport VENICE_API_KEY=...\nnpm run briefing-bot -- brief\n```\n\nTest the post path locally:\n\n```bash\nTRADEOS_BRIEFING_PLATFORM=stdout npm run briefing-bot -- post\n```\n\nPost to Discord:\n\n```bash\nexport TRADEOS_BRIEFING_PLATFORM=discord\nexport DISCORD_WEBHOOK_URL=...\nexport VENICE_API_KEY=...\nnpm run briefing-bot -- post\n```\n\nMore detail: [Market Briefing Bot](docs/market-briefing-bot.md)\n\n### Ask With The CLI\n\nFetch evidence without an LLM:\n\n```bash\nnpm run cli -- digest --limit 5\nnpm run cli -- watchlist --limit 5\n```\n\nAsk a Venice-backed question. Venice is the recommended privacy-enhanced BYOK\ndefault for self-hosted workflows, and the kit defaults to `e2ee-glm-5-1`. Get\na key from the\n[Venice AI subscription page](https://venice.ai/pricing):\n\n```bash\nexport VENICE_API_KEY=...\nnpm run cli -- ask \"What changed in crypto market stress?\"\n```\n\nSubmit feedback:\n\n```bash\nnpm run cli -- feedback \\\n  --target-id digest_123 \\\n  --target-type digest \\\n  --label useful \\\n  --note \"Clear and timely\"\n```\n\n### Add TradeOS To An MCP Host\n\nRun the local stdio MCP server from this repo:\n\n```bash\nTRADEOS_API_BASE=https://api.tradeos.tech/v1/public-intel \\\nnpm --workspace @agenticsrclab/tradeos-public-intel-mcp-server run dev\n```\n\nClaude Desktop package-style configuration:\n\n```json\n{\n  \"mcpServers\": {\n    \"tradeos-public-intel\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@agenticsrclab/tradeos-public-intel-mcp-server\"],\n      \"env\": {\n        \"TRADEOS_API_BASE\": \"https://api.tradeos.tech/v1/public-intel\",\n        \"TRADEOS_PUBLIC_INTEL_KEY\": \"<optional-app-key>\",\n        \"TRADEOS_ACCOUNT_TOKEN\": \"<optional-account-token-for-watchlists>\"\n      }\n    }\n  }\n}\n```\n\nThe MCP server is local stdio today. The reserved hosted MCP endpoint is:\n\n```text\nhttps://mcp.tradeos.tech/public-intel\n```\n\nIt becomes the zero-local-infrastructure path after the hosted HTTP MCP bridge\nis deployed.\n\nMore detail: [MCP Tools](docs/mcp-tools.md)\n\nOfficial MCP Registry metadata is prepared for the existing public distribution\nnamespace:\n\n```text\nio.github.agenticsrclab/tradeos-public-intel-mcp\n```\n\nThe registry entry publishes the local stdio package first. Hosted remote MCP\nshould be added only after `https://mcp.tradeos.tech/public-intel` is live.\n\n### Use The TypeScript SDK\n\nPackage install path after NPM publication, or when consuming a local tarball:\n\n```bash\nnpm install @tradeos/public-intel-sdk\n```\n\n```ts\nimport { TradeOSPublicIntelClient } from \"@tradeos/public-intel-sdk\";\n\nconst client = new TradeOSPublicIntelClient();\n\nconst digest = await client.getMarketDigest({ limit: 5 });\nconst snapshot = await client.getTokenWatchlistSnapshot(\"VVV\", {\n  mode: \"trader\",\n  chain: \"8453\",\n});\n\nawait client.submitDigestFeedback({\n  targetType: \"digest\",\n  targetId: \"digest_123\",\n  label: \"useful\",\n  optionalNote: \"The caveats were clear.\",\n});\n```\n\n### Use The Python SDK\n\nPackage install path after PyPI publication, or when consuming a local build:\n\n```bash\npip install tradeos-public-intel\n```\n\n```python\nfrom tradeos_public_intel import TradeOSPublicIntelClient\n\nclient = TradeOSPublicIntelClient()\ndigest = client.get_market_digest(limit=5)\nsnapshot = client.get_token_watchlist_snapshot(\"VVV\", mode=\"trader\", chain=\"8453\")\n```\n\nPython 3.11 or newer is required.\n\n### Try Saved Watchlists\n\nPublic token snapshots are keyless. Saved watchlists are user-owned state and\nrequire a signed-in TradeOS account bearer token.\n\n```bash\nexport TRADEOS_ACCOUNT_TOKEN=<signed-in-account-token>\nexport TRADEOS_PUBLIC_INTEL_KEY=<optional-builder-app-key>\n```\n\n```ts\nimport { TradeOSPublicIntelClient } from \"@tradeos/public-intel-sdk\";\n\nconst client = new TradeOSPublicIntelClient({\n  accountToken: process.env.TRADEOS_ACCOUNT_TOKEN,\n  apiKey: process.env.TRADEOS_PUBLIC_INTEL_KEY,\n});\n\nconst created = await client.createWatchlist({\n  name: \"Portfolio risk monitor\",\n  mode: \"investor\",\n});\nconst watchlistId = String(created.watchlist.watchlist_id);\n\nawait client.addWatchlistItem(watchlistId, { symbol: \"VVV\", chain: \"8453\" });\nconst state = await client.getWatchlistState(watchlistId);\n\nawait client.createWatchlistNotificationChannel(watchlistId, {\n  channelKind: \"in_app\",\n  target: \"tradeos-dashboard\",\n  minSeverity: \"watch\",\n  digestFrequency: \"realtime\",\n});\n\nawait client.triggerWatchlistDeliveries(watchlistId, {\n  channelKinds: [\"in_app\"],\n  minSeverity: \"watch\",\n});\n```\n\nThe first-party watchlist GUI uses the same API:\n[Watchlist Intelligence](https://tradeos.tech/watchlists)\n\n## Access Model\n\nBuilders and users should not need a TradeOS account just to try the public kit.\n\n| Mode | Credential | Enables |\n| --- | --- | --- |\n| Public trial | none | bounded public reads and feedback writes |\n| Builder attribution | optional `TRADEOS_PUBLIC_INTEL_KEY` | app identity, abuse controls, support, potential higher public limits |\n| User-owned watchlists | `TRADEOS_ACCOUNT_TOKEN` | saved lists, state, events, channels, delivery audit, watchlist feedback |\n| Data Intel Credit linking | TradeOS sign-in or linked session identity | starter quota and DTI credit reconciliation |\n| Builder-paid product | paid API key, contract entitlement, or x402 wallet/payment | premium resources inside the builder product |\n| User-paid agent/tool | user TradeOS entitlement or x402 payment | user brings paid access to a third-party tool |\n\nDefault flow:\n\n```text\nNo account -> try public reads -> submit feedback anonymously\nBuilder gets traction -> register app / configure optional public key\nUser wants saved state -> sign in and connect an account token\nWorkflow needs premium data -> builder or user pays through x402/API entitlement\n```\n\nTradeOS can issue public-intel app keys for signed-in, email-verified builder\naccounts. The normal path is the\n[Developer Keys](https://tradeos.tech/developer/api-keys) dashboard. The CLI can\nvalidate `TRADEOS_PUBLIC_INTEL_KEY` and can manage app keys when\n`TRADEOS_ACCOUNT_TOKEN` is set for trusted automation:\n\nFor the short key setup and scale-request path, see\n[Getting API Keys And Requesting Scale](docs/getting-api-keys-and-scale.md).\n\nPublic API quota is earned, not unlimited:\n\n| Profile | Reads/min | Reads/hour | Reads/day | Symbols/day | Path |\n| --- | ---: | ---: | ---: | ---: | --- |\n| Anonymous preview | 2 | 15 | 40 | 5 | try the kit without an account |\n| Builder baseline | 5 | 50 | 100 | 10 | app key after starter expiry |\n| Builder starter/earned | 15 | 150 | 400 | 30 | 7-day starter key or recent useful feedback |\n| Reviewed project | 30 | 300 | 800 | 60 | submit `POST /quota-requests` and get operator approval |\n\nFeedback writes are bounded too: app-key writes default to 10/minute and 100/day;\nanonymous writes default to 5/minute per IP. App keys provide attribution and\nreputation, not paid entitlement. Data Intel Credits use one common unit with\nscoped spend rules: human DTI unlocks public dashboard depth, public Ask packs,\nand read-only Review Lab access; builder feedback affects app reputation and\nquota confidence. Builders can inspect lifecycle state through\n`/feedback-activity` and `/app-feedback-status`; see\n[Data Intel Credit Loop](docs/feedback-credit-loop.md) and\n[Public Intel API](docs/public-intel-api.md). Paid/x402 access is required for\nprivate intelligence products, scale, alerts, exports, replay, automation, or\ndata rights.\n\nAt launch, each free public read counts as one read unit. Batch, history,\nexport, alert, private-intelligence, and machine-scale surfaces are paid/x402\nor entitlement-gated rather than stretched into the free public API.\n\nFree public access is best-effort promotional access: no SLA, no guaranteed\navailability, no cash value, and it may be rate-limited, changed, paused,\ndegraded, or revoked to protect TradeOS infrastructure. The free public pool is\ncapped separately so paid/x402/contract capacity can be preserved.\n\nGUI Ask TradeOS follows the same conservative loop: 3 anonymous questions, 10\nsigned-in starter questions for 7 days, and 5-question feedback packs earned\nwith DTI credits.\n\n```bash\nnpm run cli -- auth\nnpm run cli -- keys create --app-name my-public-intel-app\nnpm run cli -- keys list\nnpm run cli -- keys revoke --key-id pubkey_...\n```\n\nExisting app-key secrets are not retrievable. Create or rotate a key, copy the\none-time secret, and keep it server-side.\n\nMore detail:\n\n- [Distribution Setup Guide](docs/distribution-setup-guide.md)\n- [API Keys And Feedback Provenance](docs/api-keys-and-feedback-provenance.md)\n- [Access And Payments](docs/access-and-payments.md)\n- [Paid Boundaries](docs/paid-boundaries.md)\n- [Public Fusion Signal Cockpit Preview](docs/public-fusion-signal-cockpit-preview.md)\n\n## Feature Unlock Loop\n\nMore features come from TradeOS service access, not from hidden local code in\nthe SDK, MCP server, CLI, or bot examples.\n\n```text\n1. Builder installs the free kit.\n2. Builder ships public features using digest, candidates, watchlists, proofs, and feedback writes.\n3. End users interact with the builder product and submit structured feedback.\n4. TradeOS reconciles stable target IDs into quality signals, app reputation, and scoped credits.\n5. Human DTI can unlock temporary public dashboard depth, public Ask packs, or Review Lab passes for the user.\n6. When the workflow needs production features, the builder calls paid TradeOS/x402 surfaces.\n7. TradeOS checks entitlement or payment and returns premium data.\n8. The builder product exposes the paid feature to its customer.\n```\n\n| Feature Type | How The Builder Gets It | Example |\n| --- | --- | --- |\n| Free public features | included in this kit against `api.tradeos.tech/v1/public-intel` | digest, candidates, token snapshots, watchlist state, proofs, feedback writes |\n| Package improvements | upgrade npm/PyPI packages when TradeOS ships public tools | SDK helper, MCP tool, CLI command |\n| Earned public quota | useful app-attributed feedback or approved quota request | public read depth for real builder products |\n| Data Intel Credit depth | TradeOS account/session human DTI reconciliation | dashboard history, evidence depth, public Ask packs, and Review Lab capacity |\n| Paid machine features | x402 payment or paid API entitlement | premium market pulse, validation API, automation-safe reads |\n| Enterprise features | contract, API key, or private deployment | custom universe, bulk exports, replay datasets, support |\n\n## Architecture\n\n```text\nYour app / agent / MCP host\n        |\n        | SDK, CLI, or MCP tools\n        v\nTradeOS public-intel API\n        |\n        | bounded public evidence, source refs, caveats, stable IDs\n        v\nYour product surface\n        |\n        | optional feedback writes\n        v\nTradeOS Data Intel Credit loop\n\nOptional BYOK LLM path:\n\nYour app / CLI -> recommended Venice AI BYOK path or another OpenAI-compatible provider\n              -> answer grounded in TradeOS public evidence\n\nOptional paid marketplace LLM path:\n\nYour agent -> TradeOS AntSeed/x402 `tradeos-venice-raw-inference`\n           -> raw Venice model output, not TradeOS-grounded\n\nYour agent -> TradeOS AntSeed/x402 `tradeos-venice-intel-router`\n           -> TradeOS SKU routing + evidence packet + Venice explanation layer\n```\n\nThe hosted TradeOS Venice SKUs use `e2ee-glm-5-1` by default. Market them as\nprivacy-enhanced Venice-backed inference. Do not claim TradeOS is blind to\nprompts or evidence packets in the hosted route; TradeOS still authenticates,\nmeters, routes, and assembles evidence before forwarding to Venice.\n\nMore detail: [Architecture](docs/architecture.md)\n\n## Repository Layout\n\n- `packages/sdk-js`: TypeScript SDK.\n- `packages/sdk-python`: Python SDK.\n- `packages/mcp-server`: stdio MCP server for local agent hosts.\n- `apps/tradeos-agent-cli`: BYOK CLI for evidence reads, Venice-backed asks,\n  and feedback writes.\n- `apps/market-briefing-bot`: Discord, Telegram, and stdout market briefing bot\n  powered by TradeOS public evidence.\n- `examples`: Claude Desktop and Cursor MCP configuration examples.\n- `docs`: architecture, setup, use cases, monetization, API reference, MCP\n  tools, feedback loop, access and payments, paid boundaries, safety\n  boundaries, public Fusion preview, production readiness, and consumer E2E\n  notes.\n\nStart with the [docs index](docs/README.md).\n\n## Public API\n\nDefault production base URL:\n\n```text\nhttps://api.tradeos.tech/v1/public-intel\n```\n\nUseful public reads:\n\n```text\nGET /sources/health\nGET /watchlist-capabilities\nGET /tokens/{token_ref}/watchlist-snapshot\nGET /digest-inputs\nGET /candidates\nGET /thesis-watchlist\nGET /thesis-feedback\nGET /proofs/{public_claim_id}\n```\n\nFeedback writes:\n\n```text\nPOST /conversions       digest/evidence feedback\nPOST /claim-outcomes    public claim feedback\nPOST /thesis-outcomes   thesis feedback\nPOST /watchlists/{watchlist_id}/feedback\n```\n\nBuilder access:\n\n```text\nPOST /api-keys          create an attributed builder key\nPOST /quota-requests    request reviewed public quota or paid evaluation\n```\n\nMore detail: [Public API](docs/public-intel-api.md)\n\n## Contribute Apps, Tools, And Services\n\nThis kit is meant to be a builder surface, not just a client library. TradeOS\nwelcomes contributions that help developers turn public intelligence into\nuseful products and send structured feedback back to the intelligence loop.\n\nGood contribution lanes:\n\n| Lane | Examples |\n| --- | --- |\n| Apps and bots | Discord bot, Telegram bot, Slack bot, dashboard, newsletter worker, alert worker |\n| Agent tools | MCP tools, Claude/Cursor configs, prompt packs, local agent workflows |\n| SDK helpers | typed helpers, retry handling, examples, response formatters |\n| Services and connectors | webhook relays, scheduled workers, Notion/Airtable/Slack/Discord bridges |\n| Feedback loops | provenance adapters, human review UI, outcome label collectors |\n| Builder docs | monetization recipes, integration walkthroughs, deployment notes |\n\nThe best contributions are small, runnable, and commercially useful: they help a\nbuilder ship a workflow, keep secrets server-side, respect the safety boundary,\nand make the TradeOS feedback loop stronger.\n\nStart here: [Contributing](CONTRIBUTING.md)\n\n## Safety Boundary\n\nThis kit can produce trade/action recommendation cards, but it does not place\ntrades from TradeOS infrastructure, accept exchange credentials, custody assets,\nexpose raw private telemetry, or guarantee that a token is safe. Execution,\nallocation, approvals, and exchange keys belong in the individual operator's\nprivate self-hosted environment.\n\nSee [Safety Boundaries](docs/safety-boundaries.md).\n\n## Verified Consumer Flow\n\nBefore publish, this kit was tested from fresh tarball installs as an external\nconsumer:\n\n- CLI digest read.\n- CLI Venice-backed `ask`.\n- CLI feedback write.\n- TypeScript SDK read and feedback write.\n- stdio MCP initialize, tool listing, digest read, and feedback write.\n- Python SDK read and feedback write.\n- account signup, `VVV` watchlist state/events, in-app delivery audit,\n  unverified-email skip audit, watchlist feedback, and archive against the\n  public hosts.\n\nSee [Consumer E2E](docs/consumer-e2e.md).\n\nBefore making the repository public, run the publish gate in\n[Production Readiness](docs/production-readiness.md).\n\n## Explore TradeOS\n\n- [TradeOS app](https://tradeos.tech)\n- [TradeOS Public Intelligence Kit on GitHub](https://github.com/agenticsrclab/tradeos-public-intel-kit)\n- [Live market intelligence](https://tradeos.tech/market)\n- [Watchlist Intelligence](https://tradeos.tech/watchlists)\n- [Ask TradeOS](https://tradeos.tech/ask)\n- [Developer Keys](https://tradeos.tech/developer/api-keys)\n- [Review Lab](https://tradeos.tech/review)\n- [How TradeOS Works](https://tradeos.tech/how-it-works)\n- [Public docs](https://tradeos.tech/docs/)\n- [Public machine-readable docs](https://tradeos.tech/llms.txt)\n- [x402 discovery](https://tradeos.tech/.well-known/x402.json)\n- [x402 listings](https://tradeos.tech/x402/v1/listings)\n",
  "bytes": 34028,
  "sha": "7bb4bafcc4d929157164d8b7aea4d0788983a250a864967364c76903881048aa",
  "repo_slug": "agenticsrclab/tradeos-public-intel-kit",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_agenticsrclab_tradeos_public_i_7c90eeb9/readme"
}