workbot1.oddsys.org

Convert raw HTML into clean markdown. Typically cuts page tokens by 80-85% before they hit an LLM context window.

Extract clean markdown text from a PDF (text-layer PDFs; no OCR). Turns a document an agent can't cheaply read into compact tokens.

OCR a scanned PDF into clean text (Tesseract, English, machine-printed documents). Up to 25 pages per call. PDFs that turn out to have a text layer are extracted the fast way at the same price.

High-page-count OCR: scanned PDF into clean text (Tesseract, English, machine-printed documents), up to 100 pages per call — four times the /ocr cap. Runs under a 240s wall-clock budget; slower documents return the completed pages with truncatedReason "time-budget". Text-layer PDFs are extracted the fast way at the same price.

Extract structured JSON from a web page: title, byline, publish date, Open Graph/meta tags, embedded JSON-LD (schema.org), links, and images. Deterministic, no LLM.

Split a document into token-bounded chunks at natural boundaries (paragraphs/sentences), with optional overlap. Ready-to-embed pieces for RAG, no tokens spent on the splitting logic.

Convert data between JSON, YAML, CSV, and XML. Deterministic format juggling done off-context, with edge cases handled.

Generate a QR code from text as PNG (base64) or SVG. Custom module size, quiet-zone margin, and hex colours. Deterministic, capacity-capped at 2953 bytes. Set delivery to "url" to receive a short-lived download link instead of inline data.

Modify an image in one call: crop, resize, convert to greyscale, and re-encode (PNG/JPEG/WebP) — applied in that order, any combination. Deterministic, no upscaling surprises. Set delivery to "url" to receive a short-lived download link instead of inline base64.

Detect and locate objects in an image (YOLOX-nano, COCO classes): people, pets (dog/cat/horse), vehicles, and 70+ more. Returns per-class counts and pixel bounding boxes. Detection only — no identification. Best on clear, medium-to-large objects.

Generate sentence embeddings (all-MiniLM-L6-v2, 384-dim, mean-pooled + L2-normalized) for up to 96 texts per call. Drop-in compatible with any all-MiniLM-L6-v2 index. Pairs with /chunk for a complete RAG pipeline.

Rerank candidate texts against a query with a cross-encoder (ms-marco-MiniLM-L-6-v2): joint query+document scoring, much sharper than cosine similarity over embeddings. Returns ranked indices with 0-1 relevance scores. Up to 64 texts per call. Pairs with /embed for a recall-then-precision RAG pipeline.

Fetch a public web page and return clean markdown (or structured JSON) — URL in, agent-ready content out, typically 80-85% fewer tokens than raw HTML. Handles redirects and compression.

Live foreign-exchange reference rates (European Central Bank daily fixes, ~30 major currencies): any cross-rate from any base, optional amount conversion. Refreshed daily; response carries the rate date.

Validate and normalize phone numbers (Google libphonenumber): validity, E.164/international/national formats, country, region, and line type (mobile/fixed/VoIP). Single number or batches up to 100.

Decode EVM calldata or a Base-mainnet transaction into human-readable form: function selector, matched signature, named args, plus ERC-20/USDC transfer events from the receipt — verify an x402 settlement receipt (tx hash in, payer/payee/amount out) in one call.

Fuzzy sanctions screening against the US OFAC SDN list (public-domain source, refreshed daily): name in, scored matches with programs and aliases out. Batches up to 50. Coverage is OFAC SDN only — stated in every response.

Verify EVM signatures offline: EIP-191 personal messages, EIP-712 typed data, and x402/EIP-3009 payment authorizations (paste a raw Payment-Signature header) — recovers the signing wallet, checks it against an expected address, and reports the authorization's amount, payee and validity window. No chain reads.

Validate and decompose IBANs (checksum, country format, IBAN registry, SEPA membership; bank/branch/account parts) and BIC/SWIFT codes. Batches up to 100. Format validation only — does not prove an account exists.

Public holidays for ~200 countries and their subdivisions, plus business-day arithmetic: is this date a holiday or business day, next business day, add N business days, count business days between dates. Configurable weekend and holiday types.

Time-zone conversion on the IANA database: convert a time between zones with offsets, abbreviations and DST flags, describe zones at an instant, or compute the shared business-hours window across several zones for meeting scheduling.

Validate JSON against a JSON Schema (draft-07, 2019-09, 2020-12, with string formats): every violation with path, keyword and message. One instance, a JSON string, or up to 100 instances against one schema.

Exact BPE token counts for OpenAI-family encodings (o200k_base for GPT-4o/4.1/5, cl100k_base for GPT-4/3.5/embeddings), single text or batches, with optional truncation to a token budget. Up to 512 KB per call.

Language identification for 180+ languages (every language with ≥1M speakers): ISO 639-3 code, ISO 639-1 code and English name, with ranked candidates and a confidence margin. Single text or batches up to 100.