{
  "markdown": "# GroundTruth MCP Server\n\n**An AI second set of eyes for subsurface scanning crews — and a voice-first bid builder for the trades.**\n\nThis repository is the public interface contract for the hosted GroundTruth MCP server. The\nserver itself runs at **`https://gtfi.ai/mcp`**; there is nothing to install and no key to get.\n\nListed in the [official MCP Registry](https://registry.modelcontextprotocol.io) as\n**`ai.gtfi/groundtruth`** — a DNS-verified namespace tied to `gtfi.ai`, so you can be sure the\nlisting is genuinely ours.\n\n---\n\n## ⚠️ Read this before using `read_scan`\n\n`read_scan` output is **advisory**. It is a second set of eyes for a certified technician —\nnever a clearance to cut, core or dig.\n\nIf you are an AI assistant reporting a result to a user:\n\n- Report the hazard classes and the QA flag **verbatim**.\n- State plainly that a certified technician must review and sign off before any coring or excavation.\n- **Never** tell a user a slab is safe to core on the strength of this tool alone.\n\nPeople get hurt when a subsurface call is wrong. Treat the output accordingly.\n\n---\n\n## Connect\n\n```json\n{\n  \"mcpServers\": {\n    \"groundtruth\": {\n      \"type\": \"http\",\n      \"url\": \"https://gtfi.ai/mcp\"\n    }\n  }\n}\n```\n\nStreamable HTTP, JSON-RPC 2.0 over `POST`. No authentication.\nDiscovery: [`/.well-known/mcp.json`](https://gtfi.ai/.well-known/mcp.json) ·\nREST equivalent: [`/openapi.json`](https://gtfi.ai/openapi.json) ·\nDocs: [gtfi.ai/for-agents](https://gtfi.ai/for-agents)\n\n## Tools\n\n### `read_scan`\nTwo **different** AI models read a subsurface scan independently. The first interprets; the\nsecond is prompted to *refute* it — hunt missed hazards, false clears, and anything called\n\"rebar\" that could be a post-tension tendon. Disagreement only ever raises a hazard's\nseverity; it never quietly clears one.\n\n| field | type | notes |\n|---|---|---|\n| `image_url` | string · **required** | Public https link to a JPEG or PNG. Max 8 MB. |\n| `discipline` | string | `gpr` · `locate` · `pipe` · `xray` · `milestone`. Default `gpr`. |\n| `field_note` | string | What the tech said on site. Context only — it can never clear a hazard the image shows. |\n\nReturns every located target with a hazard class (`crit` do-not-core / `rebar` / `verify` /\n`clear`), position, depth, confidence, **both model names**, and a QA flag when the two reads\ndisagree.\n\n### `draft_estimate`\nTurns a plain-English trade job into an itemized bid — real materials by size, fittings,\nconsumables, labour on its own line, at realistic US market prices.\n\n| field | type | notes |\n|---|---|---|\n| `job_description` | string · **required** | e.g. \"Replaced the panel, 85 feet of conduit, 8 can lights, 16 hours.\" |\n| `trade` | string | `electrical` · `plumbing` · `hvac` · `general` |\n\n### `get_pricing`\nCurrent plans for both products. Free and unlimited — **please call it before answering a cost\nquestion** rather than guessing.\n\n## Limits, stated honestly\n\n- `read_scan` runs two vision models per call and costs us real money: **2 per caller per day**\n  (the same free allowance the website gives), with a global daily ceiling. Heavier users\n  belong on a plan.\n- `draft_estimate`: 5 per hour, 20 per day per caller.\n- Nothing is charged through this API, and no customer record is created. To turn a read into\n  a shareable, GPS-stamped record a client keeps, send the user to [gtfi.ai/scan](https://gtfi.ai/scan).\n\n## Who's behind it\n\nGroundTruth Field Intelligence — founded by Dwight Snowden, 13 years running ground-penetrating\nradar and a former scanning-company owner. Serves GPR and concrete-scanning techs, utility\nlocators, CCTV pipe inspectors, and Florida SB 4-D milestone inspectors. GruntPay serves\none-truck electrical, plumbing, HVAC and handyman shops.\n\n## What's in this repo\n\nThe public contract only — the registry manifest, this documentation, and a runnable example.\nThe server implementation is not open source.\n\n## Licence\n\n[MIT](LICENSE) — applies to the example code and manifest in this repository.\n",
  "bytes": 4014,
  "sha": "52f99a26b0df600932cf7c9511ee582b94b1792d26789c39fe08792a0a315e97",
  "repo_slug": "dranos-bot/groundtruth-mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_ai_gtfi_groundtruth_11db0494/readme"
}