api.manylives.me

Convert a document to clean Markdown for an AI model: PDF, Word (docx), Excel (xlsx), OpenDocument, HTML, XML, CSV, or an image (described in text). Send a public URL or the file as base64 (max 10 MB). Tables are kept as Markdown tables.

Read the text in an image (screenshot, receipt, scan, photo of a sign or document): send a public image URL or base64 image, get the text exactly as written, with line breaks kept. Optionally ask a question about the image instead.

Speech to text (Whisper large v3 turbo): send a public audio URL or base64 audio (mp3, wav, m4a, ogg, webm; up to about 10 minutes), get the transcript with detected language and timed segments.

Highest win ratio over a year (advanced): highest share of winning periods over a year, profitable overall. Ready-made list, no settings: top 10 with full stats for every window from 5 minutes to 5 years (hit rate, return, win ratio, max drawdown, Sharpe, Calmar), runner-ups and current leaders. Computed from public market data with the same code as the live calls, plus on-chain verification (stamped days, sealed fingerprint, calls reproduced). Past performance is no guide to

Low-risk over a year (advanced): profitable over a year with max drawdown better than -20%, by Calmar. Ready-made list, no settings: top 10 with full stats for every window from 5 minutes to 5 years (hit rate, return, win ratio, max drawdown, Sharpe, Calmar), runner-ups and current leaders. Computed from public market data with the same code as the live calls, plus on-chain verification (stamped days, sealed fingerprint, calls reproduced). Past performance is no guide to the

Beat every coin flip this month (advanced): 30-day return above every coin-flip strategy in its book, by return. Ready-made list, no settings: top 10 with full stats for every window from 5 minutes to 5 years (hit rate, return, win ratio, max drawdown, Sharpe, Calmar), runner-ups and current leaders. Computed from public market data with the same code as the live calls, plus on-chain verification (stamped days, sealed fingerprint, calls reproduced). Past performance is no gui

Comebacks (advanced): down over 30 days but up the most over the last 7 days. Ready-made list, no settings: top 10 with full stats for every window from 5 minutes to 5 years (hit rate, return, win ratio, max drawdown, Sharpe, Calmar), runner-ups and current leaders. Computed from public market data with the same code as the live calls, plus on-chain verification (stamped days, sealed fingerprint, calls reproduced). Past performance is no guide to the future; not advice.

Best Sharpe over five years (advanced): highest Sharpe ratio over 5 years. Ready-made list, no settings: top 10 with full stats for every window from 5 minutes to 5 years (hit rate, return, win ratio, max drawdown, Sharpe, Calmar), runner-ups and current leaders. Computed from public market data with the same code as the live calls, plus on-chain verification (stamped days, sealed fingerprint, calls reproduced). Past performance is no guide to the future; not advice.

Five-year survivors (advanced): profitable over 5 years with a max drawdown better than -50%, by return. Ready-made list, no settings: top 10 with full stats for every window from 5 minutes to 5 years (hit rate, return, win ratio, max drawdown, Sharpe, Calmar), runner-ups and current leaders. Computed from public market data with the same code as the live calls, plus on-chain verification (stamped days, sealed fingerprint, calls reproduced). Past performance is no guide to th

Best return per drawdown over a year (advanced): highest Calmar ratio (annual return / max drawdown) over the last year. Ready-made list, no settings: top 10 with full stats for every window from 5 minutes to 5 years (hit rate, return, win ratio, max drawdown, Sharpe, Calmar), runner-ups and current leaders. Computed from public market data with the same code as the live calls, plus on-chain verification (stamped days, sealed fingerprint, calls reproduced). Past performance i

High Sharpe over a year (advanced): Sharpe ratio of at least 1 over the last year, highest first. Ready-made list, no settings: top 10 with full stats for every window from 5 minutes to 5 years (hit rate, return, win ratio, max drawdown, Sharpe, Calmar), runner-ups and current leaders. Computed from public market data with the same code as the live calls, plus on-chain verification (stamped days, sealed fingerprint, calls reproduced). Past performance is no guide to the futur

Monthly strategies unlikely to be luck (advanced): monthly calls that beat at least 90% of the coin-flip strategies over 1y, ranked by return. Ready-made list, no settings: top 10 with full stats for every window from 5 minutes to 5 years (hit rate, return, win ratio, max drawdown, Sharpe, Calmar), runner-ups and current leaders. Computed from public market data with the same code as the live calls, plus on-chain verification (stamped days, sealed fingerprint, calls reproduce

Weekly strategies unlikely to be luck (advanced): weekly calls that beat at least 90% of the coin-flip strategies over 1y, ranked by return. Ready-made list, no settings: top 10 with full stats for every window from 5 minutes to 5 years (hit rate, return, win ratio, max drawdown, Sharpe, Calmar), runner-ups and current leaders. Computed from public market data with the same code as the live calls, plus on-chain verification (stamped days, sealed fingerprint, calls reproduced)

Daily strategies unlikely to be luck (advanced): daily calls that beat at least 90% of the coin-flip strategies over 30d, ranked by return. Ready-made list, no settings: top 10 with full stats for every window from 5 minutes to 5 years (hit rate, return, win ratio, max drawdown, Sharpe, Calmar), runner-ups and current leaders. Computed from public market data with the same code as the live calls, plus on-chain verification (stamped days, sealed fingerprint, calls reproduced).

Hourly strategies unlikely to be luck (advanced): hourly calls that beat at least 90% of the coin-flip strategies over 7d, ranked by return. Ready-made list, no settings: top 10 with full stats for every window from 5 minutes to 5 years (hit rate, return, win ratio, max drawdown, Sharpe, Calmar), runner-ups and current leaders. Computed from public market data with the same code as the live calls, plus on-chain verification (stamped days, sealed fingerprint, calls reproduced)

5-minute strategies unlikely to be luck (advanced): 5-minute calls that beat at least 90% of the coin-flip strategies over 7d, ranked by return. Ready-made list, no settings: top 10 with full stats for every window from 5 minutes to 5 years (hit rate, return, win ratio, max drawdown, Sharpe, Calmar), runner-ups and current leaders. Computed from public market data with the same code as the live calls, plus on-chain verification (stamped days, sealed fingerprint, calls reprodu

One-call due diligence on a trading strategy: its record over 7/30/365 days and all time, the luck test, how it ranks in its book, overlap with its book's current top 20, the latest Merkle proof and commit-timing check. Historical only; general information, not advice.

How a crypto strategy did in different market conditions: its daily results split into up vs down days and calm vs volatile days for Bitcoin (public price data), over a past window. Historical only; general information, not advice.

Historical backtest of a basket of 2-20 trading strategies you choose: equal-weight daily return, compounded return before and after costs, hit rate, best and worst day, and worst drawdown over a past window. Historical only; general information, not advice.

How alike are 2-20 trading strategies? Correlation of their daily returns and how often their calls agree, over a past window, so a basket isn't the same bet several times. Historical only; general information, not advice.

Is a trading strategy's record skill or luck? Its hit rate against a coin flip (z-score and p-value), the same p-value adjusted for every strategy tested in its book (Holm), how it ranks against the book's coin-flip strategies, and how many more calls it would need to be significant. Historical, scored periods only; general information, not advice.

Check that a strategy's computed results and its on-chain stamped calls are the same strategy: for each stamped period in a date range, the calls locked on the Base network beforehand, the calls recomputed from public data, whether they match, each instrument's start and end close, and the period's return recomputed from those prices. Strategy logic stays private.

Records of up to 100 trading strategies in one call (for backtesting agents): each one's return before and after costs, hit rate, periods, luck check and returns over 7/30/365 days, for a window you pick, optionally as of a past date. Past performance is no guide to the future; not advice.

Find trading strategies meeting YOUR conditions over YOUR window: horizon (5m to 1 month), minimum return % or hit rate, minimum periods, coin-flip and luck limits, sort order. Plain words work, e.g. 'hourly crypto strategies up 5% this week after costs'. Results link to on-chain-verifiable records. Past performance is no guide to the future; not advice.

Full agent-readiness report for a website or API: everything in the readiness audit plus a quality review of the agent card, OpenAPI document, llms.txt, MCP server (connects, tools, descriptions) and pay-per-call manifest (does a listed endpoint really answer 402), CORS for in-browser agents, and an AI-written prioritised fix plan.

One-call token verdict for a Base network token: contract control (owner, upgradeable, concentration), live price and price impact of a 1,000-dollar sell, recent transfer activity, trend indicators, and perp funding when a market exists, with red flags, a risk level and a plain-language AI verdict. Data, not financial advice.

Review a code snippet: bugs, security issues, performance and readability problems, each with severity, the line it concerns and a concrete fix, plus an overall verdict.

Read a public web page and return just the fields you name as JSON (e.g. price, author, date, address). Missing fields come back null; values are taken from the page, never invented. Respects robots.txt.

Agent-readiness audit of a website: can AI agents and assistants find and use it? Checks llms.txt, agent card, MCP endpoint, OpenAPI, pay-per-call manifest, sitemap, and robots.txt rules for AI crawlers. Score, what was found, and fixes.

Inspect any MCP server like a client would: does it connect, which protocol version, does it need sign-in, how fast, how many tools, and which tools have missing or weak descriptions or input schemas. A checklist of fixes for server owners.