{"uid":"cap_9E-tttbiCIguCTmNoEjBN","slug":"vibes-coded-json-repair-api-9be19889","name":"Vibes-Coded JSON Repair API","description":"Pay-per-call JSON for AI agents: skill search, listing optimization, install plans, productization, and artifact risk scoring. HTTP 402 → USDC. Full skill marketplace on-site.","url":"https://vibes-coded.com/api/v1/outcomes/json-repair","method":"POST","headers":{},"bodySchema":null,"responseSchema":{"type":"json","example":{"ok":true,"parsed":{"name":"agent","tools":["search","calc"],"retries":3},"confidence":0.98,"valid_json":true,"errors_fixed":["stripped ```json fence","removed trailing comma","smart quotes -> ascii"]}},"example":null,"exampleRequest":null,"tags":["x402"],"displayCostAmount":"0.01","displayCostAsset":"USDC","priceDynamic":false,"priceHint":null,"priceStatus":"priced","priceSource":"probe","requiresHandshake":false,"reviewCount":0,"rating":{"score":"0.00","successRate":"0.00","reviews":0,"stars":null,"state":"unrated"},"availabilityStatus":"unknown","priceObserved":null,"sessionDeposit":null,"pricing":{"kind":"static","summary":"$0.01/call","primary":{"kind":"static","protocol":"x402","network":"base","amountUsd":"0.01","per":"call","confidence":"exact"},"accepted":[{"kind":"static","protocol":"x402","network":"base","amountUsd":"0.01","per":"call","confidence":"exact"}]},"paymentMethods":[{"uid":"pm_HD0NkgTTCh6Fe1hwUmm02","protocol":"x402","methodType":"crypto","chain":"base","mode":"charge","costAmount":"0.01","costPer":"request","priority":0,"asset":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","unit":"request","depositMicros":null,"planRef":null}],"brandName":null,"brandSlug":null,"brandBaseUrl":null,"brandDocsUrl":null,"whatItDoes":"Repairs malformed or broken JSON strings — fixing fences, trailing commas, smart quotes, and other common errors — returning parsed output with confidence score and list of corrections made.","exampleAgentPrompt":"This JSON string from the LLM output is broken — it has a trailing comma and markdown code fences around it. Can you repair it and give me the parsed result along with what was fixed?","exampleUseCases":[{"title":"Cleaning LLM-generated JSON responses","prompt":"The AI just returned a JSON blob wrapped in triple-backtick fences with a trailing comma — it won't parse. Can you strip the fences, fix the comma, and give me the clean parsed object?"},{"title":"Fixing smart quotes in API payloads","prompt":"I have a JSON string where someone copied from a Word doc and all the quotes are curly smart quotes — can you convert them to regular ASCII quotes and parse the result so I can use it?"},{"title":"Recovering malformed agent tool output","prompt":"My agent returned this half-broken JSON: '{\"name\":\"agent\",\"tools\":[\"search\",\"calc\"],\"retries\":3,}' — the trailing comma breaks everything. Fix it and tell me your confidence that the repair is correct."}],"resultDescription":"Returns a JSON object with: `ok` (boolean success flag), `parsed` (the repaired and parsed JSON value), `confidence` (0–1 float indicating repair confidence), `valid_json` (boolean), and `errors_fixed` (array of human-readable strings describing each correction made, e.g. 'stripped ```json fence', 'removed trailing comma', 'smart quotes -> ascii').","failureModes":["Input is so corrupted it cannot be repaired — returns ok:false with confidence near 0","Input is not JSON-like at all (e.g. plain prose) — service may return low confidence or error","Network or payment failure — HTTP 402 if USDC payment not attached","Ambiguous structure (e.g. multiple possible repairs) — confidence score will be low, parsed result may be best-guess only"],"whenToPreferThis":"Choose this endpoint when you have LLM- or agent-generated JSON that fails to parse due to common formatting issues like markdown code fences, trailing commas, or smart quotes, and you need both the repaired object and an audit trail of what was changed. It is pay-per-call at $0.01 USDC via x402, making it cost-effective for occasional or high-volume JSON cleanup in automated pipelines. Prefer this over writing custom repair logic when you need confidence scoring and a structured list of corrections.","instructions":null,"reviewSummary":null,"reviewSummaryHighlights":null,"reviewSummaryConcerns":null,"reviewSummaryGeneratedAt":null,"activationCount":0,"lastUsedAt":null,"lastSuccessfullyRanAt":null,"lastHealthCheckAt":"2026-09-14T12:52:04.111Z","isFirstParty":false}