All projects matching “context”

URL-to-Markdown web extraction and context compression for AI agents. Fetch a public web page and return clean, task-relevant, citation-verifiable LLM context with exact excerpts, locators, hashes, and estimated tokens saved. Use before an agent summarizes, answers questions from, or cites a long or noisy URL. Deterministic and model-free. Public HTML only; no PDFs, logins, or JavaScript-only page

Compress long notes or conversation context into compact operational state for agent handoffs

GET /api/v1/postcode/:postcode - Single UK Postcode Socio-Economic Profile (ONS Census 2021 demographics, IMD 2019 deciles, Land Registry prices, GBP/sqm, EPC ratings, broadband speeds)

Find transcript passages matching a phrase or terms in a public YouTube video, with timestamp citations.

Extract and validate structured expense data from receipt text or a provider-neutral receipt extraction. Use when an agent needs merchant, date, tax, total, line items, confidence, duplicate detection, and arithmetic validation before creating or reconciling an expense. The result is advisory and always requires human review.

SwapHunt — a crypto market-context toolkit for trading agents: live prices, market regime, volatility, sentiment, funding, macro events, seasonality and anti-FOMO sanity checks, all pay-per-call. This route returns a single asset's live spot price with 24h context: price, 24h change %, 24h high/low and volume — one clean JSON snapshot (BTC, ETH, SOL, any major). For many assets at once use /price/

Linkrra Synap: Google Gemma 4 26B A4B Instruct (model "gemma-4-26b-a4b-it") - fast general model, long context. OpenAI-compatible chat completions for this one model, 128K context, served only on dedicated GPUs we operate, 4-bit AWQ. Pay per call in USDC on Base via x402 - no account, no API key. From $0.001 per call, usage-priced per token above that (GET /v1/models). Streaming supported. Product

Global Pro adds independent higher-timeframe context, SMA family, regression channel, anchored VWAP, Fibonacci, confirmed pivots, RSI and ATR evidence to the same 520-candle chart and aligned Spot plan. Context follows the selected timeframe; conditional paths and unclear Elliott counts are explicit. Free full-report recovery. Market or Limit orders require separate wallet review and signing.

[MCP: optimize_context] Pay-per-call JSON context pruning & LLM token budget enforcement. Saves up to 85% prompt tokens.

This resource provides editorial reference material on commercial solar-control window film for the Los Angeles context. It supports topic research and preparation of questions for a qualified project reviewer. Read the returned material for its specific coverage, source context, and limitations.

Statically analyse a Model Context Protocol (MCP) server manifest for security risks. Tool descriptions enter the model context, a prime prompt-injection vector. Deterministic and offline (inline manifest). Flags dangerous capabilities (command/code execution, filesystem, credential and network access), injection patterns, hidden unicode and unconstrained parameters. Returns severity-ranked findin

Agent-orchestrator Solana wallet-intelligence brief via x402: 24h/7d context, cutoffs, coverage gaps, and next evidence depth. Use to choose the next Scry route — not as a substitute for a known-wallet dossier, Birdeye charts, or Nansen smart-money. Evidence routing only.

Scrape website page to Markdown. Fetch any public web page URL and get clean, LLM-ready text or Markdown: main content, code blocks, tables, links and headings, for RAG and agent context. Static HTML, Markdown or text; optional section ID. Default 20,000 output characters, max 100,000; input 2 MiB. No JavaScript rendering (see /extract/rendered), logins or paywalls. Empty pages, bad sections, brow

Compile 1-8 documents into a token-budgeted context pack, then measure exact retention of 1-32 caller-labelled evidence spans. Returns the pack, source-linked passages, input and output hashes, original and compiled token estimates, tokensSaved, source coverage, and requiredEvidenceRetentionPercent. Retention is exact span matching only: not accuracy, answer quality, verification, or hallucination

Cross-class positioning panels per asset — futures funding stress, options 1M risk-reversal skew, order-book imbalance, and 7-day taker buy/sell ratio. Each panel: current z, percentile, 30-day sparkline, one-line narrative. Response carries a _next block with active-signals / thesis / regime-context / top-setups / peer-cluster so an agent can go from positioning read to per-asset conviction, hist

Routes requests across open-weight models using task fit, tool calling, vision, JSON format, context, output capacity, provider availability, and price. Send POST application/json with model set to ai-open-models/auto and a non-empty messages array. task_type is optional; when omitted, Auto classifies the request. Supports function tools, JSON responses, and mixed text/image messages using public

Returns licensed ENTSO-E and SMARD spot-market context for compute, DePIN, battery and power-trading agents: published day-ahead prices, volatility, load, renewable share when available, the observed share of published negative-price hours, uncertainty, freshness and expiry.

Deterministic prompt-injection firewall and honeypot scrubber for AI agents. Sanitizes untrusted text and blocks detected wallet-draining commands before context ingestion.

For an investment theme (e.g. HBM, 유리기판, 원전, robotics), the Korean listed companies linked to it with evidence graded confirmed / circumstantial / expectation: DART filing snippets with receipt numbers, regional revenue, earnings and order headlines, and price context since the signal. Separates real revenue from theme hype.

One call, many sources: search Hacker News, GitHub, Stack Overflow, and Wikipedia by keyword and get back a unified, ranked result set — titles, URLs, scores/stars, and snippets — across all of them at once. The general research/search read agents make to ground answers, find code, surface discussion, and gather context. No API keys, no per-source accounts, free upstreams. Pass `sources` to narrow

Verify an onchain payment or investigate a failed Base transaction in one call with transaction, receipt, success status, confirmations, gas and fee context.

Send messy text (web page, product listing, article, docs, log, spec) plus the fields you want; get one valid JSON object with exactly those keys, typed values, nulls for anything missing, or an array of records. Offload parsing and save your own context tokens. Run by an AI agent.

Scrape any web page: URL to Markdown / HTML to Markdown for LLM context. Readability article extraction strips nav, ads, scripts, and footers while keeping tables and code blocks. Optional render:true executes JavaScript in a headless browser for SPAs (React/Next/Vue).

What can I use instead of buttermilk, self-raising flour, creme fraiche or sour cream? Get head chef Cristopher's verified swaps: yes, no or depends (and on what), with amount and context where given, including swaps not to make. Also fresh herbs, butter, oil, sugar, flour, fresh tomatoes, fresh stock, whole spices; others get a 404 listing them. Refused, never guessed: allergens and dairy-, egg-

News prediction market catalyst API for agents. Ranks prediction market signals by freshness, confidence, urgency, liquidity, and probability context.

Use when a page will not read correctly with a plain fetch — a single-page app, content loaded by JavaScript, or markup so noisy the text is buried. Fetches it with a headless browser (rendering JavaScript) and returns the main readable content as clean Markdown — ideal to feed an LLM as context. JSON body { url } (public http/https), optional include_images (default false). Returns title, final_u

Esports match schedule for esports and esports-betting agents. Returns today's matches with team context and storylines across LoL, Valorant, and CS2.

Requires an unauthenticated caller-owned baseline; returns real current NS/MX records, added/removed sets and tri-state target-provider transition. Current DNS comes from one bounded native recursive-resolver query; existing Dossier provides context. Resolver caching may apply; no account ownership or global migration-completion claim. Existing provider patterns are reused without new invented att

Is this payment address safe to pay? DEEP version — deep counterparty assessment: same shape as /api/assess but scans more on-chain settlement history for higher confidence. POST {target:{payto_address}, payment_context?, policy?}. Returns risk_score 0-100, allow/review/block, confidence, per-dimension signals, key_factors and a recommended_policy. Use when the payment is large enough that a thin-

WhiteMagic hosted recall — agent memory for cross-session continuity: read-only hybrid search (BM25 + vector + associative) over a curated public corpus. Call when an agent needs durable context from prior sessions or the public knowledge store. Keyless discovery (initialize/tools/list) is free; free evaluation keys and a daily no-key trial are available. One paid tool call per payment.