[{"_1":2,"_3":-5,"_4":-5},"loaderData",{"_5":6,"_7":8,"_9":10},"actionData","errors","root",{"_224":225,"_226":227},"routes/_app.x.$handle.$name",{"_96":95,"_32":45,"_104":86,"_27":105,"_13":14,"_106":-5,"_11":107,"_108":109,"_110":111,"_112":113,"_114":115,"_20":-5,"_116":-5,"_117":-5,"_118":25,"_119":120,"_121":58,"_122":-5,"_123":-5,"_124":125},"routes/_app.x.$handle.$name.branch.$branch.version.$version.diff",{"_11":12,"_13":14,"_15":16,"_17":-5,"_18":-5},"project",{"_43":44,"_32":45,"_38":46,"_47":-5,"_48":25,"_49":50,"_51":50,"_52":53,"_54":-5,"_18":55,"_56":31,"_57":58,"_59":-5,"_60":-5,"_61":62,"_63":64,"_65":-5,"_66":-5,"_67":-5,"_68":-5,"_69":70,"_71":70,"_72":73,"_74":73,"_75":73,"_76":70,"_77":70,"_78":79,"_80":81,"_82":83,"_84":25},"version",20,"changes",[19],"aiChatInfo","userId",{"_20":21,"_22":23,"_24":25,"_26":25,"_27":28,"_29":28},"file",{"_30":31,"_32":28,"_33":34,"_35":37,"_38":39,"_40":14,"_41":42,"_24":25},"previous",{"_30":31,"_32":28,"_33":34,"_35":36},"deleted",false,"deletedHere","path","main.ts","latestPath","fileId","b0887434-56a7-11f1-b5c3-ee650bb23af1","name","fileType","http","content","import { Hono } from \"npm:hono\";\nimport { paymentMiddleware } from \"npm:x402-hono\";\nimport { jsonrepair } from \"npm:jsonrepair\";\nimport Ajv from \"npm:ajv\";\nimport OpenAI from \"npm:openai\";\nimport { parseVal } from \"https://esm.town/v/std/utils/index.ts\";\n\n// ---------------------------------------------------------------------------\n// Config\n// ---------------------------------------------------------------------------\nconst PAY_TO = (Deno.env.get(\"PAY_TO_ADDRESS\") ??\n \"0x3F11bAab8B0454131B6c81684524ccF1c98b4968\") as `0x${string}`;\nconst PRICE_USD = Deno.env.get(\"PRICE_USD\") ?? \"$0.01\";\n// x402-hono@1.2.0 only accepts legacy network names (\"base\", \"base-sepolia\").\n// CAIP-2 strings (eip155:8453) need x402-hono v2 (unreleased as of 2026-05).\n// Until then, /repair emits a v1 challenge — x402scan flags this but it still settles.\nconst X402_NETWORK = (Deno.env.get(\"X402_NETWORK\") ?? \"base\") as \"base\" | \"base-sepolia\";\n// PayAI: open, free, Base mainnet exact-USDC + built-in bazaar discovery\n// (auto-indexes services with discoverable:true at /discovery/resources).\n// Alternatives: https://facilitator.xpay.sh · https://facilitator.daydreams.systems\nconst FACILITATOR_URL = Deno.env.get(\"X402_FACILITATOR_URL\") ?? \"https://facilitator.payai.network\";\nconst LLM_MODEL = \"Qwen/Qwen3.6-35B-A3B:wafer\";\nconst MAX_INPUT = 100_000;\n\nconst openai = new OpenAI({\n baseURL: \"https://nano-gpt.com/api/v1\",\n apiKey: Deno.env.get(\"NANOGPT_API_KEY\"),\n});\n\nconst LLM_SYSTEM_PROMPT = `You are a JSON repair engine. Return ONLY a single valid JSON value: no prose, no explanation, no markdown fences. If a JSON Schema is provided, the output must validate against it exactly. Fix syntax errors: unbalanced brackets, trailing commas, unquoted or single-quoted keys and strings, missing commas, comments, and Python-style True, False, None. Preserve the original data and intent. Never invent values that are not present in or clearly derivable from the input. For a required field with no recoverable value, use null. Coerce types only when intent is unambiguous. Always return valid JSON, best effort.`;\n\n// ---------------------------------------------------------------------------\n// Helpers\n// ---------------------------------------------------------------------------\nconst ajv = new Ajv({ allErrors: true, strict: false });\n\nfunction validate(data: unknown, schema?: Record): { valid: boolean; errors: unknown[] } {\n if (!schema) return { valid: true, errors: [] };\n const fn = ajv.compile(schema);\n const valid = fn(data) as boolean;\n return { valid, errors: valid ? [] : (fn.errors ?? []) };\n}\n\nfunction stripFences(s: string): string {\n return s.replace(/^```(?:json)?\\s*\\n?/i, \"\").replace(/\\n?```\\s*$/i, \"\").trim();\n}\n\nfunction tryParse(s: string): { ok: boolean; data?: unknown } {\n try {\n return { ok: true, data: JSON.parse(s) };\n } catch {\n return { ok: false };\n }\n}\n\nfunction success(method: string, data: unknown, valid: boolean, errors: unknown[], model: string | null) {\n return Response.json(\n { ok: true, method, valid, data, validationErrors: errors, model },\n { status: 200 },\n );\n}\n\nfunction fail(status: number, error: string) {\n return Response.json({ ok: false, error }, { status });\n}\n\n// ---------------------------------------------------------------------------\n// App\n// ---------------------------------------------------------------------------\nconst app = new Hono();\napp.onError((err) => Promise.reject(err));\n\n// x402 payment gate — only POST /repair\napp.use(\n \"/repair\",\n paymentMiddleware(\n PAY_TO,\n {\n \"/repair\": {\n price: PRICE_USD,\n network: X402_NETWORK,\n config: {\n description:\n \"Repair malformed JSON via a fast deterministic parser, with an LLM fallback for harder cases or strict JSON Schema conformance. Returns valid JSON. Zero data retention.\",\n mimeType: \"application/json\",\n discoverable: true,\n outputSchema: {\n input: {\n type: \"http\",\n method: \"POST\",\n discoverable: true,\n bodyType: \"json\",\n schema: {\n type: \"object\",\n required: [\"input\"],\n properties: {\n input: {\n description:\n \"Malformed JSON as a string, OR any JSON value to validate against `schema`. Required. Max 100,000 characters.\",\n },\n schema: {\n type: \"object\",\n description:\n \"Optional JSON Schema (Ajv draft-07+). When provided, the repaired output is validated; deterministic results that fail validation fall through to the LLM.\",\n },\n instructions: {\n type: \"string\",\n description:\n \"Optional natural-language hint passed to the LLM fallback. Example: 'coerce dates to ISO 8601' or 'use null for unknown fields'.\",\n },\n },\n },\n },\n output: {\n type: \"json\",\n schema: {\n type: \"object\",\n properties: {\n ok: { type: \"boolean\", description: \"True on successful repair.\" },\n method: {\n type: \"string\",\n enum: [\"deterministic\", \"llm\"],\n description: \"Which repair stage produced the result.\",\n },\n valid: {\n type: \"boolean\",\n description: \"Whether the repaired data validates against the provided schema (true when no schema given).\",\n },\n data: { description: \"Repaired JSON value.\" },\n validationErrors: {\n type: \"array\",\n description: \"Ajv errors when validation fails. Empty otherwise.\",\n },\n model: {\n type: [\"string\", \"null\"],\n description: \"LLM model id when method=llm, null otherwise.\",\n },\n },\n },\n },\n },\n },\n },\n },\n { url: FACILITATOR_URL },\n ),\n);\n\n// ---------------------------------------------------------------------------\n// POST /repair\n// ---------------------------------------------------------------------------\napp.post(\"/repair\", async (c) => {\n let body: Record;\n try {\n body = await c.req.json();\n } catch {\n return fail(400, \"Request body must be valid JSON\");\n }\n\n const { input, schema, instructions } = body as {\n input?: unknown;\n schema?: Record;\n instructions?: string;\n };\n\n if (input === undefined || input === null) {\n return fail(400, \"\\\"input\\\" is required\");\n }\n\n const raw = typeof input === \"string\" ? input : JSON.stringify(input);\n if (raw.length > MAX_INPUT) {\n return fail(400, `Input exceeds ${MAX_INPUT} character limit`);\n }\n\n // --- Stage 1: already valid JSON value? ------------------------------------\n if (typeof input !== \"string\") {\n const v = validate(input, schema);\n if (v.valid) return success(\"deterministic\", input, true, [], null);\n // carry to stage 3\n }\n\n // --- Stage 2: jsonrepair on string input -----------------------------------\n let repaired: unknown | undefined;\n if (typeof input === \"string\") {\n try {\n const fixed = jsonrepair(input);\n const parsed = JSON.parse(fixed);\n const v = validate(parsed, schema);\n if (!schema || v.valid) return success(\"deterministic\", parsed, v.valid, v.errors, null);\n repaired = parsed; // carry parsed to stage 3\n } catch {\n // jsonrepair or parse failed — carry raw string to stage 3\n }\n }\n\n // --- Stage 3: LLM ---------------------------------------------------------\n try {\n const userParts: string[] = [`Input:\\n${raw}`];\n if (schema) userParts.push(`JSON Schema:\\n${JSON.stringify(schema)}`);\n if (instructions) userParts.push(`Instructions:\\n${instructions}`);\n\n const controller = new AbortController();\n const timeout = setTimeout(() => controller.abort(), 25_000);\n\n let completion: OpenAI.Chat.ChatCompletion;\n try {\n completion = await openai.chat.completions.create(\n {\n model: LLM_MODEL,\n temperature: 0,\n messages: [\n { role: \"system\", content: LLM_SYSTEM_PROMPT },\n { role: \"user\", content: userParts.join(\"\\n\\n\") },\n ],\n },\n { signal: controller.signal },\n );\n } finally {\n clearTimeout(timeout);\n }\n\n const llmOut = completion.choices?.[0]?.message?.content ?? \"\";\n const cleaned = stripFences(llmOut);\n\n let data: unknown;\n // Try direct parse first, fall back to jsonrepair\n const directParse = tryParse(cleaned);\n if (directParse.ok) {\n data = directParse.data;\n } else {\n try {\n data = JSON.parse(jsonrepair(cleaned));\n } catch {\n return fail(500, \"LLM output could not be parsed as JSON\");\n }\n }\n\n const v = validate(data, schema);\n return success(\"llm\", data, v.valid, v.errors, LLM_MODEL);\n } catch (err: unknown) {\n const msg = err instanceof Error ? err.message : String(err);\n if (msg.includes(\"abort\")) return fail(500, \"LLM request timed out (25 s)\");\n return fail(500, `LLM error: ${msg}`);\n }\n});\n\n// ---------------------------------------------------------------------------\n// GET /health\n// ---------------------------------------------------------------------------\napp.get(\"/health\", (c) => c.json({ ok: true }));\n\n// ---------------------------------------------------------------------------\n// GET /openapi.json\n// ---------------------------------------------------------------------------\napp.get(\"/openapi.json\", (c) => {\n const origin = new URL(c.req.url).origin;\n const price = PRICE_USD;\n const amount = price.replace(/^\\$/, \"\");\n\n const doc = {\n openapi: \"3.1.0\",\n info: {\n title: \"jsonfix\",\n version: \"1.0.0\",\n description: `Paid API that repairs malformed JSON. Price: ${price} USDC per request via x402. Fully ZDR, no data retention.`,\n \"x-guidance\":\n \"jsonfix repairs malformed JSON. POST a JSON body to /repair with { input: , schema?: , instructions?: }. The server runs a deterministic parser first; if that fails (or its result violates the provided schema), it falls back to a Wafer-routed Qwen LLM. The response is { ok, method: 'deterministic'|'llm', valid, data, validationErrors, model }. Each call costs \" +\n price +\n \" USDC, settled via x402 on Base mainnet. No API key, no signup. Zero data retention: nothing is stored or logged.\",\n },\n servers: [{ url: origin }],\n paths: {\n \"/repair\": {\n post: {\n summary: \"Repair malformed JSON\",\n operationId: \"repair\",\n description: `Attempts deterministic repair via jsonrepair, then falls back to an LLM. Payment of ${price} USDC required via x402 on Base mainnet.`,\n \"x-payment-info\": {\n price: { mode: \"fixed\", currency: \"USD\", amount },\n protocols: [{ x402: {} }],\n },\n requestBody: {\n required: true,\n content: {\n \"application/json\": {\n schema: {\n type: \"object\",\n required: [\"input\"],\n properties: {\n input: {\n description: \"Malformed JSON string or any JSON value\",\n },\n schema: {\n type: \"object\",\n description: \"Optional JSON Schema to validate against\",\n },\n instructions: {\n type: \"string\",\n description: \"Optional natural-language hint for the LLM\",\n },\n },\n },\n },\n },\n },\n responses: {\n \"200\": {\n description: \"Repair succeeded (best-effort)\",\n content: {\n \"application/json\": {\n schema: {\n type: \"object\",\n properties: {\n ok: { type: \"boolean\", const: true },\n method: { type: \"string\", enum: [\"deterministic\", \"llm\"] },\n valid: { type: \"boolean\" },\n data: { description: \"Repaired JSON value\" },\n validationErrors: { type: \"array\" },\n model: { type: [\"string\", \"null\"] },\n },\n },\n },\n },\n },\n \"400\": { description: \"Bad request\" },\n \"402\": { description: \"Payment required (x402)\" },\n \"500\": { description: \"Internal / LLM failure\" },\n },\n },\n },\n \"/health\": {\n get: {\n summary: \"Health check\",\n security: [],\n responses: {\n \"200\": {\n description: \"Service is healthy\",\n content: {\n \"application/json\": {\n schema: {\n type: \"object\",\n properties: { ok: { type: \"boolean\", const: true } },\n },\n },\n },\n },\n },\n },\n },\n },\n };\n\n return c.json(doc);\n});\n\n// ---------------------------------------------------------------------------\n// GET /favicon.svg\n// ---------------------------------------------------------------------------\nconst FAVICON_SVG = ``;\napp.get(\"/favicon.svg\", (c) => c.body(FAVICON_SVG, 200, { \"content-type\": \"image/svg+xml; charset=utf-8\" }));\n\n// ---------------------------------------------------------------------------\n// GET / — landing page\n// ---------------------------------------------------------------------------\napp.get(\"/\", (c) => {\n const origin = new URL(c.req.url).origin;\n const price = PRICE_USD;\n const { links } = parseVal();\n\n const html = `\n\n\n \n \n jsonfix — repair malformed JSON\n \n \n \n \n \n \n \n \n\n\n

\n\n

\n \n \n jsonfix\n \n

\n \n Live · Base mainnet\n

\n\n

\n\n \n

HTTP API · x402 USDC

\n Repair malformed
\n JSON.\n

\n Deterministic parser first. Wafer-routed Qwen fallback when bytes are too broken. One paid endpoint, schema-validated output.\n

\n \n

\n\n \n

Quick start

\n POST /repair\n

# pay-per-request via x402 wallet\ncurl -X POST ${origin}/repair \\\\\n  -H \"Content-Type: application/json\" \\\\\n  -d '{\"input\": \"{name: \\\\\"Alice\\\\\", age: 30,}\"}'

\n\n \n

For your agent

\n \n

When you produce JSON that fails to parse or fails its schema, POST { input, schema?, instructions? } to ${origin}/repair (paid: ${price} USDC via x402 on Base mainnet) and use the returned data. Free probe: GET ${origin}/openapi.json.

\n \n

\n\n \n

Request

{\n  \"input\": \"<malformed JSON or any value>\",  // required\n  \"schema\": { ... },              // optional JSON Schema\n  \"instructions\": \"...\"            // optional hint\n}

Response

{\n  \"ok\": true,\n  \"method\": \"deterministic\" | \"llm\",\n  \"valid\": true,\n  \"data\": <repaired JSON>,\n  \"validationErrors\": [],\n  \"model\": \"${LLM_MODEL}\" | null\n}

\n\n \n

\n \n \n \n

Zero data retention

\n Submitted JSON is processed in memory and discarded with the response. Nothing stored, nothing logged. Inference is routed through Wafer, a fully ZDR provider.\n

\n\n \n \n\n

\n\n`;\n\n return c.html(html);\n});\n\n// ---------------------------------------------------------------------------\n// Source link\n// ---------------------------------------------------------------------------\napp.get(\"/source\", (c) => c.redirect(parseVal().links.self.val));\n\nexport default app.fetch;\n","import { Hono } from \"npm:hono\";\nimport { paymentMiddleware } from \"npm:x402-hono\";\nimport { jsonrepair } from \"npm:jsonrepair\";\nimport Ajv from \"npm:ajv\";\nimport OpenAI from \"npm:openai\";\nimport { parseVal } from \"https://esm.town/v/std/utils/index.ts\";\n\n// ---------------------------------------------------------------------------\n// Config\n// ---------------------------------------------------------------------------\nconst PAY_TO = (Deno.env.get(\"PAY_TO_ADDRESS\") ??\n \"0x3F11bAab8B0454131B6c81684524ccF1c98b4968\") as `0x${string}`;\nconst PRICE_USD = Deno.env.get(\"PRICE_USD\") ?? \"$0.01\";\n// x402-hono@1.2.0 only accepts legacy network names (\"base\", \"base-sepolia\").\n// CAIP-2 strings (eip155:8453) need x402-hono v2 (unreleased as of 2026-05).\n// Until then, /repair emits a v1 challenge — x402scan flags this but it still settles.\nconst X402_NETWORK = (Deno.env.get(\"X402_NETWORK\") ?? \"base\") as \"base\" | \"base-sepolia\";\n// PayAI: open, free, Base mainnet exact-USDC + built-in bazaar discovery\n// (auto-indexes services with discoverable:true at /discovery/resources).\n// Alternatives: https://facilitator.xpay.sh · https://facilitator.daydreams.systems\nconst FACILITATOR_URL = Deno.env.get(\"X402_FACILITATOR_URL\") ?? \"https://facilitator.payai.network\";\nconst LLM_MODEL = \"Qwen/Qwen3.6-35B-A3B:wafer\";\nconst MAX_INPUT = 100_000;\n\nconst openai = new OpenAI({\n baseURL: \"https://nano-gpt.com/api/v1\",\n apiKey: Deno.env.get(\"NANOGPT_API_KEY\"),\n});\n\nconst LLM_SYSTEM_PROMPT = `You are a JSON repair engine. Return ONLY a single valid JSON value: no prose, no explanation, no markdown fences. If a JSON Schema is provided, the output must validate against it exactly. Fix syntax errors: unbalanced brackets, trailing commas, unquoted or single-quoted keys and strings, missing commas, comments, and Python-style True, False, None. Preserve the original data and intent. Never invent values that are not present in or clearly derivable from the input. For a required field with no recoverable value, use null. Coerce types only when intent is unambiguous. Always return valid JSON, best effort.`;\n\n// ---------------------------------------------------------------------------\n// Helpers\n// ---------------------------------------------------------------------------\nconst ajv = new Ajv({ allErrors: true, strict: false });\n\nfunction validate(data: unknown, schema?: Record): { valid: boolean; errors: unknown[] } {\n if (!schema) return { valid: true, errors: [] };\n const fn = ajv.compile(schema);\n const valid = fn(data) as boolean;\n return { valid, errors: valid ? [] : (fn.errors ?? []) };\n}\n\nfunction stripFences(s: string): string {\n return s.replace(/^```(?:json)?\\s*\\n?/i, \"\").replace(/\\n?```\\s*$/i, \"\").trim();\n}\n\nfunction tryParse(s: string): { ok: boolean; data?: unknown } {\n try {\n return { ok: true, data: JSON.parse(s) };\n } catch {\n return { ok: false };\n }\n}\n\nfunction success(method: string, data: unknown, valid: boolean, errors: unknown[], model: string | null) {\n return Response.json(\n { ok: true, method, valid, data, validationErrors: errors, model },\n { status: 200 },\n );\n}\n\nfunction fail(status: number, error: string) {\n return Response.json({ ok: false, error }, { status });\n}\n\n// ---------------------------------------------------------------------------\n// App\n// ---------------------------------------------------------------------------\nconst app = new Hono();\napp.onError((err) => Promise.reject(err));\n\n// x402 payment gate — only POST /repair\napp.use(\n \"/repair\",\n paymentMiddleware(\n PAY_TO,\n {\n \"/repair\": {\n price: PRICE_USD,\n network: X402_NETWORK,\n config: {\n description:\n \"Repair malformed JSON via a fast deterministic parser, with an LLM fallback for harder cases or strict JSON Schema conformance. Returns valid JSON. Zero data retention.\",\n mimeType: \"application/json\",\n discoverable: true,\n outputSchema: {\n input: {\n type: \"http\",\n method: \"POST\",\n discoverable: true,\n bodyType: \"json\",\n schema: {\n type: \"object\",\n required: [\"input\"],\n properties: {\n input: {\n description:\n \"Malformed JSON as a string, OR any JSON value to validate against `schema`. Required. Max 100,000 characters.\",\n },\n schema: {\n type: \"object\",\n description:\n \"Optional JSON Schema (Ajv draft-07+). When provided, the repaired output is validated; deterministic results that fail validation fall through to the LLM.\",\n },\n instructions: {\n type: \"string\",\n description:\n \"Optional natural-language hint passed to the LLM fallback. Example: 'coerce dates to ISO 8601' or 'use null for unknown fields'.\",\n },\n },\n },\n },\n output: {\n type: \"json\",\n schema: {\n type: \"object\",\n properties: {\n ok: { type: \"boolean\", description: \"True on successful repair.\" },\n method: {\n type: \"string\",\n enum: [\"deterministic\", \"llm\"],\n description: \"Which repair stage produced the result.\",\n },\n valid: {\n type: \"boolean\",\n description: \"Whether the repaired data validates against the provided schema (true when no schema given).\",\n },\n data: { description: \"Repaired JSON value.\" },\n validationErrors: {\n type: \"array\",\n description: \"Ajv errors when validation fails. Empty otherwise.\",\n },\n model: {\n type: [\"string\", \"null\"],\n description: \"LLM model id when method=llm, null otherwise.\",\n },\n },\n },\n },\n },\n },\n },\n },\n { url: FACILITATOR_URL },\n ),\n);\n\n// ---------------------------------------------------------------------------\n// POST /repair\n// ---------------------------------------------------------------------------\napp.post(\"/repair\", async (c) => {\n let body: Record;\n try {\n body = await c.req.json();\n } catch {\n return fail(400, \"Request body must be valid JSON\");\n }\n\n const { input, schema, instructions } = body as {\n input?: unknown;\n schema?: Record;\n instructions?: string;\n };\n\n if (input === undefined || input === null) {\n return fail(400, \"\\\"input\\\" is required\");\n }\n\n const raw = typeof input === \"string\" ? input : JSON.stringify(input);\n if (raw.length > MAX_INPUT) {\n return fail(400, `Input exceeds ${MAX_INPUT} character limit`);\n }\n\n // --- Stage 1: already valid JSON value? ------------------------------------\n if (typeof input !== \"string\") {\n const v = validate(input, schema);\n if (v.valid) return success(\"deterministic\", input, true, [], null);\n // carry to stage 3\n }\n\n // --- Stage 2: jsonrepair on string input -----------------------------------\n let repaired: unknown | undefined;\n if (typeof input === \"string\") {\n try {\n const fixed = jsonrepair(input);\n const parsed = JSON.parse(fixed);\n const v = validate(parsed, schema);\n if (!schema || v.valid) return success(\"deterministic\", parsed, v.valid, v.errors, null);\n repaired = parsed; // carry parsed to stage 3\n } catch {\n // jsonrepair or parse failed — carry raw string to stage 3\n }\n }\n\n // --- Stage 3: LLM ---------------------------------------------------------\n try {\n const userParts: string[] = [`Input:\\n${raw}`];\n if (schema) userParts.push(`JSON Schema:\\n${JSON.stringify(schema)}`);\n if (instructions) userParts.push(`Instructions:\\n${instructions}`);\n\n const controller = new AbortController();\n const timeout = setTimeout(() => controller.abort(), 25_000);\n\n let completion: OpenAI.Chat.ChatCompletion;\n try {\n completion = await openai.chat.completions.create(\n {\n model: LLM_MODEL,\n temperature: 0,\n messages: [\n { role: \"system\", content: LLM_SYSTEM_PROMPT },\n { role: \"user\", content: userParts.join(\"\\n\\n\") },\n ],\n },\n { signal: controller.signal },\n );\n } finally {\n clearTimeout(timeout);\n }\n\n const llmOut = completion.choices?.[0]?.message?.content ?? \"\";\n const cleaned = stripFences(llmOut);\n\n let data: unknown;\n // Try direct parse first, fall back to jsonrepair\n const directParse = tryParse(cleaned);\n if (directParse.ok) {\n data = directParse.data;\n } else {\n try {\n data = JSON.parse(jsonrepair(cleaned));\n } catch {\n return fail(500, \"LLM output could not be parsed as JSON\");\n }\n }\n\n const v = validate(data, schema);\n return success(\"llm\", data, v.valid, v.errors, LLM_MODEL);\n } catch (err: unknown) {\n const msg = err instanceof Error ? err.message : String(err);\n if (msg.includes(\"abort\")) return fail(500, \"LLM request timed out (25 s)\");\n return fail(500, `LLM error: ${msg}`);\n }\n});\n\n// ---------------------------------------------------------------------------\n// GET /health\n// ---------------------------------------------------------------------------\napp.get(\"/health\", (c) => c.json({ ok: true }));\n\n// ---------------------------------------------------------------------------\n// GET /openapi.json\n// ---------------------------------------------------------------------------\napp.get(\"/openapi.json\", (c) => {\n const origin = new URL(c.req.url).origin;\n const price = PRICE_USD;\n const amount = price.replace(/^\\$/, \"\");\n\n const doc = {\n openapi: \"3.1.0\",\n info: {\n title: \"jsonfix\",\n version: \"1.0.0\",\n description: `Paid API that repairs malformed JSON. Price: ${price} USDC per request via x402. Fully ZDR, no data retention.`,\n \"x-guidance\":\n \"jsonfix repairs malformed JSON. POST a JSON body to /repair with { input: , schema?: , instructions?: }. The server runs a deterministic parser first; if that fails (or its result violates the provided schema), it falls back to a Wafer-routed Qwen LLM. The response is { ok, method: 'deterministic'|'llm', valid, data, validationErrors, model }. Each call costs \" +\n price +\n \" USDC, settled via x402 on Base mainnet. No API key, no signup. Zero data retention: nothing is stored or logged.\",\n },\n servers: [{ url: origin }],\n paths: {\n \"/repair\": {\n post: {\n summary: \"Repair malformed JSON\",\n operationId: \"repair\",\n description: `Attempts deterministic repair via jsonrepair, then falls back to an LLM. Payment of ${price} USDC required via x402 on Base mainnet.`,\n \"x-payment-info\": {\n price: { mode: \"fixed\", currency: \"USD\", amount },\n protocols: [{ x402: {} }],\n },\n requestBody: {\n required: true,\n content: {\n \"application/json\": {\n schema: {\n type: \"object\",\n required: [\"input\"],\n properties: {\n input: {\n description: \"Malformed JSON string or any JSON value\",\n },\n schema: {\n type: \"object\",\n description: \"Optional JSON Schema to validate against\",\n },\n instructions: {\n type: \"string\",\n description: \"Optional natural-language hint for the LLM\",\n },\n },\n },\n },\n },\n },\n responses: {\n \"200\": {\n description: \"Repair succeeded (best-effort)\",\n content: {\n \"application/json\": {\n schema: {\n type: \"object\",\n properties: {\n ok: { type: \"boolean\", const: true },\n method: { type: \"string\", enum: [\"deterministic\", \"llm\"] },\n valid: { type: \"boolean\" },\n data: { description: \"Repaired JSON value\" },\n validationErrors: { type: \"array\" },\n model: { type: [\"string\", \"null\"] },\n },\n },\n },\n },\n },\n \"400\": { description: \"Bad request\" },\n \"402\": { description: \"Payment required (x402)\" },\n \"500\": { description: \"Internal / LLM failure\" },\n },\n },\n },\n \"/health\": {\n get: {\n summary: \"Health check\",\n security: [],\n responses: {\n \"200\": {\n description: \"Service is healthy\",\n content: {\n \"application/json\": {\n schema: {\n type: \"object\",\n properties: { ok: { type: \"boolean\", const: true } },\n },\n },\n },\n },\n },\n },\n },\n },\n };\n\n return c.json(doc);\n});\n\n// ---------------------------------------------------------------------------\n// GET /favicon.svg\n// ---------------------------------------------------------------------------\nconst FAVICON_SVG = ``;\napp.get(\"/favicon.svg\", (c) => c.body(FAVICON_SVG, 200, { \"content-type\": \"image/svg+xml; charset=utf-8\" }));\n\n// ---------------------------------------------------------------------------\n// GET / — landing page\n// ---------------------------------------------------------------------------\napp.get(\"/\", (c) => {\n const origin = new URL(c.req.url).origin;\n const price = PRICE_USD;\n const { links } = parseVal();\n\n const html = `\n\n\n \n \n jsonfix — repair malformed JSON\n \n \n \n \n \n \n \n \n\n\n

\n\n

\n \n \n jsonfix\n \n

\n \n Live · Base mainnet\n

\n\n

\n\n \n

HTTP API · x402 USDC

\n Repair malformed
\n JSON.\n

\n Deterministic parser first. Wafer-routed Qwen fallback when bytes are too broken. One paid endpoint, schema-validated output.\n

\n \n

\n\n \n

Quick start

\n POST /repair\n

# pay-per-request via x402 wallet\ncurl -X POST ${origin}/repair \\\\\n  -H \"Content-Type: application/json\" \\\\\n  -d '{\"input\": \"{name: \\\\\"Alice\\\\\", age: 30,}\"}'

\n\n \n

For your agent

\n \n

When you produce JSON that fails to parse or fails its schema, POST { input, schema?, instructions? } to ${origin}/repair (paid: ${price} USDC via x402 on Base mainnet) and use the returned data. Free probe: GET ${origin}/openapi.json.

\n \n

\n\n \n

Request

{\n  \"input\": \"<malformed JSON or any value>\",  // required\n  \"schema\": { ... },              // optional JSON Schema\n  \"instructions\": \"...\"            // optional hint\n}

Response

{\n  \"ok\": true,\n  \"method\": \"deterministic\" | \"llm\",\n  \"valid\": true,\n  \"data\": <repaired JSON>,\n  \"validationErrors\": [],\n  \"model\": \"${LLM_MODEL}\" | null\n}

\n\n \n

\n \n \n \n

Zero data retention

\n Submitted JSON is processed in memory and discarded with the response. Nothing stored, nothing logged. Inference is routed through Wafer, a fully ZDR provider.\n

\n\n \n \n\n

\n\n`;\n\n return c.html(html);\n});\n\n// ---------------------------------------------------------------------------\n// Source link\n// ---------------------------------------------------------------------------\napp.get(\"/source\", (c) => c.redirect(parseVal().links.self.val));\n\n// ---------------------------------------------------------------------------\n// GET /.well-known/x402 — machine-readable discovery doc\n// ---------------------------------------------------------------------------\napp.get(\"/.well-known/x402\", (c) => {\n const origin = new URL(c.req.url).origin;\n const amount = PRICE_USD.replace(/^\\$/, \"\");\n return c.json({\n x402Version: 1,\n name: \"jsonfix\",\n description:\n \"Repair malformed JSON via deterministic parser + LLM fallback. Pay-per-call USDC on Base mainnet. Zero data retention.\",\n payTo: PAY_TO,\n resources: [\n {\n method: \"POST\",\n path: \"/repair\",\n resource: `${origin}/repair`,\n description:\n \"Repair malformed JSON. Body: { input, schema?, instructions? }. Returns { ok, method, valid, data, validationErrors, model }.\",\n accepts: [\n {\n scheme: \"exact\",\n network: X402_NETWORK,\n payTo: PAY_TO,\n asset: \"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913\",\n mimeType: \"application/json\",\n price: { mode: \"fixed\", currency: \"USD\", amount },\n maxTimeoutSeconds: 300,\n extra: { name: \"USD Coin\", version: \"2\" },\n },\n ],\n inputSchema: {\n type: \"object\",\n required: [\"input\"],\n properties: {\n input: { description: \"Malformed JSON string or any JSON value. Max 100,000 chars.\" },\n schema: { type: \"object\", description: \"Optional JSON Schema to validate against.\" },\n instructions: { type: \"string\", description: \"Optional natural-language hint for the LLM.\" },\n },\n },\n outputSchema: {\n type: \"object\",\n properties: {\n ok: { type: \"boolean\" },\n method: { type: \"string\", enum: [\"deterministic\", \"llm\"] },\n valid: { type: \"boolean\" },\n data: { description: \"Repaired JSON value.\" },\n validationErrors: { type: \"array\" },\n model: { type: [\"string\", \"null\"] },\n },\n },\n },\n ],\n docs: `${origin}/openapi.json`,\n landing: origin,\n });\n});\n\n// ---------------------------------------------------------------------------\n// GET /llms.txt — agent-facing usage doc (plain text)\n// ---------------------------------------------------------------------------\napp.get(\"/llms.txt\", (c) => {\n const origin = new URL(c.req.url).origin;\n const body = `# jsonfix\n\nPaid HTTP API that repairs malformed JSON. ${PRICE_USD} USDC per call on Base mainnet via x402. No API key, no signup. Zero data retention — nothing stored or logged.\n\n## When to use\n\nCall this whenever you produce JSON that fails to parse, or fails its JSON Schema.\n\n## Endpoint\n\nPOST ${origin}/repair\nContent-Type: application/json\nBody: { \"input\": , \"schema\"?: , \"instructions\"?: }\n\n## Response\n\n{\n \"ok\": true,\n \"method\": \"deterministic\" | \"llm\",\n \"valid\": true,\n \"data\": ,\n \"validationErrors\": [],\n \"model\": \"Qwen/Qwen3.6-35B-A3B:wafer\" | null\n}\n\n## Pipeline\n\n1. Already-valid JSON value → returned unchanged.\n2. \\`jsonrepair\\` deterministic pass — fixes unbalanced brackets, trailing commas, single/unquoted keys, Python literals.\n3. LLM fallback (Qwen via NanoGPT Wafer, fully ZDR) — handles harder cases and strict schema conformance.\n\n## Payment\n\nx402 challenge on first call. Sign a USDC EIP-3009 \\`transferWithAuthorization\\` for ${PRICE_USD} to ${PAY_TO} on Base mainnet, return as \\`X-PAYMENT\\` header. Use any x402 client (x402-fetch, x402-axios, agentcash CLI).\n\n## Discovery\n\n- OpenAPI: ${origin}/openapi.json\n- x402 well-known: ${origin}/.well-known/x402\n- Health: ${origin}/health\n`;\n return c.body(body, 200, { \"content-type\": \"text/plain; charset=utf-8\" });\n});\n\nexport default app.fetch;\n","createdAt",["D",1779718951887],"branchVersion","branchId","b085f60a-56a7-11f1-b5c3-ee650bb23af1","id","b0853166-56a7-11f1-b5c3-ee650bb23af1","jsonfix",["D",1779541364623],"deletedAt","legacyEnvVars","privacy","public","httpPrivacy","description","Paid API that repairs malformed JSON via x402","imageUrl","adfedcbc-bb7b-4109-9ae2-91aad81c49ac","pinnedFileId","pinPreview",true,"pinMigrated","screenshotVersion","likeCount",0,"orgId","071ebdee-2ec6-407d-a65d-5c20110f98a5","tursoName","tursoHostname","tursoMeasuredAt","pinned","userPerm","r","valPerm","blobPerm","rw","emailPerm","sqlitePerm","projectPerm","telemetryPerm","user",{"_96":95,"_103":99},"org",{"_96":95,"_97":-5,"_98":99,"_18":55,"_100":101,"_102":25},"branch",{"_13":85,"_32":86,"_43":42,"_87":44,"_88":-5,"_89":-5,"_90":-5,"_91":-5,"_92":-5,"_93":45,"_94":95},"isOwned",23,"main","projectId","forkedBranchId","forkedBranchVersion","forkedProjectPrivacy","forkedProjectId","forkedProjectOrgId","projectName","userHandle","ngmi","handle","displayName","avatar","https://img.clerk.com/eyJ0eXBlIjoicHJveHkiLCJzcmMiOiJodHRwczovL2ltYWdlcy5jbGVyay5kZXYvb2F1dGhfZ2l0aHViL2ltZ18ya1RTUlZRSG9oYkxSUXlrVjF2TWRVS0lIWDQifQ","tier","free","canRestrictVals","profileImageUrl","branchName","","adoptOrgId",{"_43":44,"_32":45,"_38":220,"_47":-5,"_48":25,"_49":50,"_51":50,"_52":53,"_54":-5,"_18":55,"_56":31,"_57":58,"_59":-5,"_60":-5,"_61":62,"_63":64,"_65":-5,"_66":-5,"_67":-5,"_68":-5,"_69":70,"_71":70,"_72":73,"_74":73,"_75":73,"_76":70,"_77":70,"_78":221,"_80":222,"_82":223,"_84":25},"tree",[185,187,188,189,190,191],"files",[185,186,187,188,189,190,191],"counts",{"_174":62,"_175":176,"_177":62,"_178":179,"_180":181},"contents",[126,127,128,129,130,131],"upstreamProject","upstreamBranch","emptyEnvVars","flags",[],"hasHttpFiles","grants","orgs","customDomains",{},{"_43":167,"_32":168,"_30":169,"_133":-5,"_41":42,"_40":170,"_134":58,"_38":171,"_13":62,"_33":172,"_35":-5,"_24":25,"_137":-5,"_27":173},{"_43":160,"_32":161,"_30":162,"_133":-5,"_41":42,"_40":163,"_134":58,"_38":164,"_13":62,"_33":20,"_35":165,"_24":25,"_137":-5,"_27":166},{"_43":153,"_32":154,"_30":155,"_133":-5,"_41":42,"_40":156,"_134":58,"_38":157,"_13":62,"_33":20,"_35":158,"_24":25,"_137":-5,"_27":159},{"_43":146,"_32":147,"_30":148,"_133":-5,"_41":42,"_40":149,"_134":58,"_38":150,"_13":62,"_33":20,"_35":151,"_24":25,"_137":-5,"_27":152},{"_43":139,"_32":140,"_30":141,"_133":-5,"_41":42,"_40":142,"_134":58,"_38":143,"_13":62,"_33":20,"_35":144,"_24":25,"_137":-5,"_27":145},{"_43":132,"_32":28,"_30":31,"_133":-5,"_41":42,"_40":14,"_134":25,"_38":135,"_13":136,"_33":34,"_35":37,"_24":25,"_137":-5,"_27":138},"2ac8f522-5845-11f1-a863-ee650bb23af1","parentId","latest",["D",1779718951887],10,"label",[28],"b3eba9ee-5831-11f1-96d9-ee650bb23af1","deno.json","b3ebaa8e-5831-11f1-96d9-ee650bb23af1",5,["D",1779710592033],"{\n \"$schema\": \"https://raw.githubusercontent.com/denoland/deno/98f62cee78e85bfc47c62ed703777c6bc8794f1c/cli/schemas/config-file.v1.json\",\n \"lock\": false,\n \"compilerOptions\": {\n \"noImplicitAny\": false,\n \"strict\": true,\n \"types\": [\"https://www.val.town/types/valtown.d.ts\"],\n \"lib\": [\n \"dom\",\n \"dom.iterable\",\n \"dom.asynciterable\",\n \"deno.ns\",\n \"deno.unstable\"\n ]\n },\n \"lint\": {\n \"include\": [\"deno:/https/esm.town/**/*\"],\n \"rules\": {\n \"exclude\": [\"no-explicit-any\"]\n }\n },\n \"node_modules_dir\": false,\n \"experimental\": {\n \"unstable-node-globals\": true,\n \"unstable-temporal\": true,\n \"unstable-worker-options\": true,\n \"unstable-sloppy-imports\": true\n }\n}",[140],"b40ab9ba-5831-11f1-a3d4-ee650bb23af1","CLAUDE.md","b40aba82-5831-11f1-a3d4-ee650bb23af1",8,["D",1779710592229],"# CLAUDE.md\n\nGuide Claude Code (claude.ai/code) for this repo.\n\n## Project\n\nSingle-file Val Town HTTP val (`main.ts`) — paid JSON-repair API. Runs on Deno in Val Town serverless. No build, no tests, no `package.json`. Deps = URL imports (`npm:`, `https://esm.town`, `https://esm.sh`) resolved at runtime.\n\nEntry: `export default app.fetch` at bottom of `main.ts`. Val Town calls per HTTP request.\n\n## Workflow (vt CLI)\n\nVal Town working copy via `vt` CLI. **Not** git repo. State in `.vt/state.json` (val id, branch, version).\n\n```bash\nvt status # working-tree status vs. remote\nvt push # upload local changes to Val Town\nvt pull # fetch remote changes\nvt watch # auto-sync on save\nvt tail # stream logs from deployed val\nvt browse # open val on val.town in browser\nvt branch # list/delete branches\nvt checkout # switch branch\n```\n\nNo local dev server — code runs only after push. Edit-test loop = `vt watch` + `vt tail`. Test live endpoint with curl on val URL.\n\nType-check local: `deno check main.ts` (uses `deno.json` with val.town types).\n\n## Architecture\n\nThree-stage repair pipeline in `POST /repair` (main.ts:83):\n\n1. **Stage 1 — already-valid passthrough** (main.ts:107). If `input` is non-string JSON value, validates vs optional `schema`, return as-is with `method: \"deterministic\"`.\n2. **Stage 2 — `jsonrepair`** (main.ts:115). If `input` is string, run `jsonrepair` then `JSON.parse`. Success → return with `method: \"deterministic\"`. Schema given + result invalid → *fall through to stage 3* (don't return).\n3. **Stage 3 — LLM** (main.ts:128). Call `gpt-4o-mini` (Val Town `std/openai`, no API key) with `LLM_SYSTEM_PROMPT` + raw input + optional schema + optional instructions. Strip markdown fences, try `JSON.parse`, fall back to `jsonrepair(cleaned)`. Validate, return with `method: \"llm\"`. 25s `AbortController` timeout.\n\nValidation = Ajv (`allErrors: true, strict: false`). `validate()` no-op when no schema. Stage 2 never falls through if no schema — parsed repair returned immediately.\n\n### Payment gate (x402)\n\n`paymentMiddleware` from `x402-hono` gates **only** `POST /repair` (main.ts:62). Charges `PRICE_USD` (default `$0.003`) USDC on Base mainnet to `PAY_TO`, via x402 facilitator. Health, OpenAPI, landing, source = free.\n\n### Endpoints\n\n- `POST /repair` — paid, body `{ input, schema?, instructions? }`. `input` required; max 100k chars.\n- `GET /health` — `{ ok: true }`.\n- `GET /openapi.json` — generated OpenAPI 3.1 doc (origin from request URL).\n- `GET /` — Tailwind/Twind landing page with usage examples.\n- `GET /source` — 302 to val source page on val.town.\n\nErrors: `app.onError` rethrows so Val Town logs full stacks (main.ts:59). Inside `/repair`, business errors return `{ ok: false, error }` via `fail()`.\n\n## Environment variables\n\n- `PAY_TO_ADDRESS` — recipient address (default hardcoded).\n- `PRICE_USD` — string like `\"$0.003\"`.\n- `X402_FACILITATOR_URL` — defaults to `https://facilitator.x402.rs`.\n\nOpenAI key not needed — `std/openai` wired to Val Town billed account.\n\n## Val Town conventions\n\nFull platform guide in `AGENTS.md`. Key constraints:\n\n- Pin import versions (e.g. `npm:jsonrepair`, `https://esm.town/v/std/utils@85-main/index.ts`).\n- Never use Deno KV; use `std/blob` or `std/sqlite` if persistence added.\n- Use `Deno.env.get()` for secrets; never bake in.\n- Server-only utils like `parseVal()` must not appear in client code — inject results into HTML.\n- Redirects: use `new Response(null, { status: 302, headers: { Location } })` — `Response.redirect` broken on Val Town.\n- React work (none now) needs `?deps=react@18.2.0,react-dom@18.2.0` + `/** @jsxImportSource https://esm.sh/react@18.2.0 */` header.",[147],"b43b8da6-5831-11f1-af50-ee650bb23af1","AGENTS.md","b43b8eb4-5831-11f1-af50-ee650bb23af1",9,["D",1779710592287],"You are an advanced assistant specialized in generating Val Town code.\n\n## Core Guidelines\n\n- Ask clarifying questions if requirements ambiguous\n- Give complete functional solutions, not skeletons\n- Test logic against edge cases before final\n- All code must follow Val Town platform rules\n- If section too complex, refactor into subcomponents\n\n## Code Standards\n\n- Generate TypeScript or TSX\n- Add TypeScript types/interfaces for all data structures\n- Prefer official SDKs over raw API calls\n- Ask user for API/library docs if unsure\n- **Never bake secrets into code** — use env vars\n- Comment complex logic only (skip obvious)\n- Use modern ES6+ and functional style where possible\n\n## Types of triggers\n\n### 1. HTTP Trigger\n\n- Web APIs and endpoints\n- Handle HTTP req/res\n- Example:\n\n```ts\nexport default async function (req: Request) {\n return new Response(\"Hello World\");\n}\n```\n\nHTTP trigger files have http in name like `foobar.http.tsx`\n\n### 2. Cron Triggers\n\n- Run on schedule\n- Cron expressions for timing\n- Example:\n\n```ts\nexport default async function () {\n // Scheduled task code\n}\n```\n\nCron trigger files have cron in name like `foobar.cron.tsx`\n\n### 3. Email Triggers\n\n- Process incoming emails\n- Email workflows\n- Example:\n\n```ts\nexport default async function (email: Email) {\n // Process email\n}\n```\n\nEmail trigger files have email in name like `foobar.email.tsx`\n\n\n## Val Town Standard Libraries\n\nVal Town gives hosted services and utility functions.\n\n### Blob Storage\n\n```ts\nimport { blob } from \"https://esm.town/v/std/blob\";\nawait blob.setJSON(\"myKey\", { hello: \"world\" });\nlet blobDemo = await blob.getJSON(\"myKey\");\nlet appKeys = await blob.list(\"app_\");\nawait blob.delete(\"myKey\");\n```\n\n### SQLite\n\n```ts\nimport { sqlite } from \"https://esm.town/v/stevekrouse/sqlite\";\nconst TABLE_NAME = 'todo_app_users_2';\n// Create table - do this before usage and change table name when modifying schema\nawait sqlite.execute(`CREATE TABLE IF NOT EXISTS ${TABLE_NAME} (\n id INTEGER PRIMARY KEY AUTOINCREMENT,\n name TEXT NOT NULL\n)`);\n// Query data\nconst result = await sqlite.execute(`SELECT * FROM ${TABLE_NAME} WHERE id = ?`, [1]);\n```\n\nNote: Changing SQLite schema → rename table (e.g., add _2 or _3) for fresh table.\n\n### OpenAI\n\n```ts\nimport { OpenAI } from \"https://esm.town/v/std/openai\";\nconst openai = new OpenAI();\nconst completion = await openai.chat.completions.create({\n messages: [\n { role: \"user\", content: \"Say hello in a creative way\" },\n ],\n model: \"gpt-4o-mini\",\n max_tokens: 30,\n});\n```\n\n### Email\n\n```ts\nimport { email } from \"https://esm.town/v/std/email\";\n// By default emails the owner of the val\nawait email({ \n subject: \"Hi\", \n text: \"Hi\", \n html: \"

Hi

\"\n});\n```\n\n## Val Town Utility Functions\n\nVal Town gives utility functions for common project tasks.\n\n### Importing Utilities\n\nAlways pin version to avoid breaking changes:\n\n```ts\nimport { parseProject, readFile, serveFile } from \"https://esm.town/v/std/utils@85-main/index.ts\";\n```\n\n### Available Utilities\n\n\n#### **serveFile** - Serve project files with proper content types\n\nExample in Hono:\n\n```ts\n// serve all files in frontend/ and shared/\napp.get(\"/frontend/*\", c => serveFile(c.req.path, import.meta.url));\napp.get(\"/shared/*\", c => serveFile(c.req.path, import.meta.url));\n```\n\n#### **readFile** - Read files from within the project:\n\n```ts\n// Read a file from the project\nconst fileContent = await readFile(\"/frontend/index.html\", import.meta.url);\n```\n\n#### **listFiles** - List all files in the project\n\n```ts\nconst files = await listFiles(import.meta.url);\n```\n\n#### **parseProject** - Extract information about the current project from import.meta.url\n\nUseful for linking back to val, e.g. \"view source\" urls:\n\n```ts\nconst projectVal = parseProject(import.meta.url);\nconsole.log(projectVal.username); // Owner of the project\nconsole.log(projectVal.name); // Project name\nconsole.log(projectVal.version); // Version number\nconsole.log(projectVal.branch); // Branch name\nconsole.log(projectVal.links.self.project); // URL to the project page\n```\n\n*Critical*: `parseProject` and other Standard Library utilities ONLY RUN ON SERVER.\nFor client access, run on server and pass to client via HTML splice or API request.\n\n## Val Town Platform Specifics\n\n- **Redirects:** Use `return new Response(null, { status: 302, headers: { Location: \"/place/to/redirect\" }})` — `Response.redirect` broken\n- **Images:** Avoid external/base64 images. Use emojis, unicode, or icon fonts/libraries\n- **AI Image:** Inline generate via: `

`\n- **Storage:** DO NOT use Deno KV module for storage\n- **Browser APIs:** DO NOT use `alert()`, `prompt()`, `confirm()`\n- **Weather Data:** Use open-meteo (no API keys) unless specified otherwise\n- **View Source:** Add view source link via `import.meta.url.replace(\"ems.sh\", \"val.town)\"` (or pass to client) + `target=\"_top\"`\n- **Error Debugging:** Add `` to HTML for client-side errors\n- **Error Handling:** Only try...catch with clear local resolution; avoid catches that just log or return 500s. Let errors bubble up with full context\n- **Environment Variables:** Use `Deno.env.get('keyname')` when needed; prefer key-free APIs\n- **Imports:** Use `https://esm.sh` for npm/Deno deps — works server + browser\n- **Storage Strategy:** Only use backend storage if required; prefer static client-side sites\n- **React Configuration:** Pin versions with `?deps=react@18.2.0,react-dom@18.2.0` and start file with `/** @jsxImportSource https://esm.sh/react@18.2.0 */`\n- Pin all React deps and sub-deps to same version\n- **Styling:** Default TailwindCSS via `` unless specified\n\n## Project Structure and Design Patterns\n\n### Recommended Directory Structure\n```\n├── backend/\n│ ├── database/\n│ │ ├── migrations.ts # Schema definitions\n│ │ ├── queries.ts # DB query functions\n│ │ └── README.md\n│ └── routes/ # Route modules\n│ ├── [route].ts\n│ └── static.ts # Static file serving\n│ ├── index.ts # Main entry point\n│ └── README.md\n├── frontend/\n│ ├── components/\n│ │ ├── App.tsx\n│ │ └── [Component].tsx\n│ ├── favicon.svg\n│ ├── index.html # Main HTML template\n│ ├── index.tsx # Frontend JS entry point\n│ ├── README.md\n│ └── style.css\n├── README.md\n└── shared/\n ├── README.md\n └── utils.ts # Shared types and functions\n```\n\n### Backend (Hono) Best Practices\n\n- Hono is the recommended API framework\n- Main entry point should be `backend/index.ts`\n- **Static asset serving:** Use the utility functions to read and serve project files:\n ```ts\n import { readFile, serveFile } from \"https://esm.town/v/std/utils@85-main/index.ts\";\n \n // serve all files in frontend/ and shared/\n app.get(\"/frontend/*\", c => serveFile(c.req.path, import.meta.url));\n app.get(\"/shared/*\", c => serveFile(c.req.path, import.meta.url));\n \n // For index.html, often you'll want to bootstrap with initial data\n app.get(\"/\", async c => {\n let html = await readFile(\"/frontend/index.html\", import.meta.url);\n \n // Inject data to avoid extra round-trips\n const initialData = await fetchInitialData();\n const dataScript = ``;\n \n html = html.replace(\"\", `${dataScript}`);\n return c.html(html);\n });\n ```\n- Create RESTful API routes for CRUD operations\n- Always include this snippet at the top-level Hono app to re-throwing errors to see full stack traces:\n ```ts\n // Unwrap Hono errors to see original error details\n app.onError((err, c) => {\n throw err;\n });\n ```\n\n### Database Patterns\n- Run migrations on startup or comment out for performance\n- Change table names when modifying schemas rather than altering\n- Export clear query functions with proper TypeScript typing\n\n## Common Gotchas and Solutions\n\n1. **Environment Limitations:** \n - Val Town runs on Deno in a serverless context, not Node.js\n - Code in `shared/` must work in both frontend and backend environments\n - Cannot use `Deno` keyword in shared code\n - Use `https://esm.sh` for imports that work in both environments\n\n2. **SQLite Peculiarities:**\n - Limited support for ALTER TABLE operations\n - Create new tables with updated schemas and copy data when needed\n - Always run table creation before querying\n\n3. **React Configuration:**\n - All React dependencies must be pinned to 18.2.0\n - Always include `@jsxImportSource https://esm.sh/react@18.2.0` at the top of React files\n - Rendering issues often come from mismatched React versions\n\n4. **File Handling:**\n - Val Town only supports text files, not binary\n - Use the provided utilities to read files across branches and forks\n - For files in the project, use `readFile` helpers\n\n5. **API Design:**\n - `fetch` handler is the entry point for HTTP vals\n - Run the Hono app with `export default app.fetch // This is the entry point for HTTP vals`",[154],"b3fce1e6-5831-11f1-a1d8-ee650bb23af1",".vtignore","b3fce2f4-5831-11f1-a1d8-ee650bb23af1",7,["D",1779710592123],".git\n.vscode\n.cursorrules\n.DS_Store\nnode_modules\nvendor\njsonfix-test\n.env\n.env.local\n*.original.md",[161],"b3c47d9c-5831-11f1-96d9-ee650bb23af1",".claude","b3c47e78-5831-11f1-96d9-ee650bb23af1",4,["D",1779710591756],"directory",[168],"pulls","branches",1,"forks","envVars",{"_182":176,"_184":62},"connections",{"_182":62,"_183":62},"count","unfilledCount","emptyCount",{"_43":167,"_32":168,"_30":169,"_133":-5,"_41":42,"_40":170,"_134":58,"_38":217,"_13":62,"_33":172,"_35":-5,"_24":25,"_137":-5,"_193":218,"_27":219},{"_43":208,"_32":209,"_30":210,"_133":169,"_41":42,"_40":211,"_134":58,"_38":212,"_13":213,"_33":20,"_35":214,"_24":25,"_137":-5,"_193":215,"_27":216},{"_43":160,"_32":161,"_30":162,"_133":-5,"_41":42,"_40":163,"_134":58,"_38":205,"_13":62,"_33":20,"_35":165,"_24":25,"_137":-5,"_193":206,"_27":207},{"_43":153,"_32":154,"_30":155,"_133":-5,"_41":42,"_40":156,"_134":58,"_38":202,"_13":62,"_33":20,"_35":158,"_24":25,"_137":-5,"_193":203,"_27":204},{"_43":146,"_32":147,"_30":148,"_133":-5,"_41":42,"_40":149,"_134":58,"_38":199,"_13":62,"_33":20,"_35":151,"_24":25,"_137":-5,"_193":200,"_27":201},{"_43":139,"_32":140,"_30":141,"_133":-5,"_41":42,"_40":142,"_134":58,"_38":196,"_13":62,"_33":20,"_35":144,"_24":25,"_137":-5,"_193":197,"_27":198},{"_43":132,"_32":28,"_30":31,"_133":-5,"_41":42,"_40":14,"_134":25,"_38":192,"_13":136,"_33":34,"_35":37,"_24":25,"_137":-5,"_193":194,"_27":195},["D",1779718951887],"children",[],[28],["D",1779710592033],[],[140],["D",1779710592229],[],[147],["D",1779710592287],[],[154],["D",1779710592123],[],[161],"91c0d942-5837-11f1-aab0-ee650bb23af1","settings.local.json","b3f45fd0-5831-11f1-889a-ee650bb23af1",18,["D",1779713111674],3,"{\n \"permissions\": {\n \"allow\": [\n \"Read(//home/christian/.claude/plugins/cache/caveman/caveman/63a91ecadbf4/skills/caveman-compress/**)\",\n \"Bash(python3 -m scripts /home/christian/dev/val/jsonfix/CLAUDE.md)\",\n \"Bash(python3 -m scripts /home/christian/dev/val/jsonfix/AGENTS.md)\",\n \"Bash(bun --version)\",\n \"Bash(mkdir -p /home/christian/dev/val/jsonfix/jsonfix-test)\",\n \"Bash(bun init *)\",\n \"Bash(curl -s https://facilitator.x402.rs/supported)\",\n \"WebSearch\",\n \"Bash(curl -s --max-time 10 https://facilitator.xpay.sh/supported)\",\n \"Bash(curl -s --max-time 10 https://facilitator.payai.network/supported)\",\n \"Bash(curl -s --max-time 10 https://facilitator.daydreams.systems/supported)\",\n \"Bash(curl -s --max-time 8 -o /dev/null -w \\\"%{http_code}\\\\\\\\n\\\" https://facilitator.xpay.sh/bazaar)\",\n \"Bash(curl -s --max-time 8 -o /dev/null -w \\\"%{http_code}\\\\\\\\n\\\" https://facilitator.xpay.sh/registry)\",\n \"Bash(curl -s --max-time 8 -o /dev/null -w \\\"%{http_code}\\\\\\\\n\\\" https://facilitator.xpay.sh/list)\",\n \"Bash(curl -s --max-time 8 -o /dev/null -w \\\"%{http_code}\\\\\\\\n\\\" https://facilitator.xpay.sh/discover)\",\n \"Bash(curl -s --max-time 8 https://facilitator.xpay.sh/)\",\n \"Bash(curl -s --max-time 10 https://facilitator.payai.network/bazaar)\",\n \"Bash(curl -s --max-time 10 https://facilitator.payai.network/discovery)\",\n \"Bash(curl -s --max-time 10 https://facilitator.payai.network/list)\",\n \"Bash(curl -s --max-time 15 https://facilitator.payai.network/discovery/resources)\",\n \"WebFetch(domain:x402scan.com)\",\n \"Bash(grep -oE '\\\"[^\\\"]{40,}\\\"' /home/christian/.claude/projects/-home-christian-dev-val-jsonfix/218d5afb-3b2c-48ef-91ee-181c48952281/tool-results/bhorjdrza.txt)\"\n ]\n }\n}\n",[],[168,209],["D",1779710591756],[186],[168],["D",1779541364623],{"_96":95,"_103":99},{"_96":95,"_97":-5,"_98":99,"_18":55,"_100":101,"_102":25},{"_13":85,"_32":86,"_43":42,"_87":44,"_88":-5,"_89":-5,"_90":-5,"_91":-5,"_92":-5,"_93":45,"_94":95},"ctx",{"_78":-5,"_123":257,"_258":259,"_260":261,"_262":25,"_263":25,"_264":25,"_265":58,"_266":-5,"_267":-7,"_268":-7,"_269":62,"_270":25,"_271":25,"_272":-5,"_273":25,"_274":25,"_275":25,"_276":-5},"ENV",{"_228":229,"_230":231,"_232":86,"_233":234,"_235":236,"_237":238,"_239":240,"_241":242,"_243":244,"_245":246,"_247":248,"_249":250,"_251":252,"_253":254,"_255":256},"REACT_APP_SENTRY_DSN","https://586020d86f05489db63a554690acf1e9@o1398359.ingest.sentry.io/6724499","RENDER_GIT_COMMIT","e87eb04e707cf64e7e532c8e8ec395f209e17f15","RENDER_GIT_BRANCH","POSTHOG_API_TOKEN","phc_nmoyuAkStEZuUp69rYlg1DZgrIt19Gx0Y4Hwf8tGgJP","POSTHOG_API_HOST","https://wart.val.town","WEB_API_TEMPLATE","https://{val}.web.val.run","APP_URL","https://www.val.town","MODULE_URL","https://esm.town","VANITY_WEB_API_TEMPLATE","https://{label}.val.run/","DENOLS_URL","https://lsp.val.town","NODE_ENV","production","CLOUDFLARE_TURNSTILE_SITE_KEY","0x4AAAAAACtaIZJH7s9rBMZ6","GOOGLE_CLIENT_ID","875989696899-jtsqkhlbdv5jrlsk8631p7ipjg3c6o62.apps.googleusercontent.com","STRIPE_PUBLISHABLE_KEY","pk_live_51N2zvWEpZTTLipIkVEwXrUG9AT0yeOOEYxszarZT1jpsZV58Ly6wlg3SltZpHN26CUalOBmUSY7adQRQiSkCZz2500qSXNkf8Y","API_URL","https://api.val.town",[],"preferencesKeybindings","default","preferencesTheme","light","preferencesFixers","preferencesCopilot","preferencesSelectionAi","preferencesCloseBrackets","preferencesChatPanelState","preferencesDprint","preferencesDenofmt","notificationCount","flagDenols","isBusinessOrgMember","businessOrgHandle","isBusinessOrgAdmin","isHigherTownieTier","navCollapsed","currentOrgId"]

\n Repair malformed\n JSON.\n

Quick start

For your agent

Request

Response

Zero data retention

\n Repair malformed\n JSON.\n

Quick start

For your agent

Request

Response

Zero data retention

Hi

\n Repair malformed
\n JSON.\n

\n Repair malformed
\n JSON.\n

Diff · v20