{"server":"io.github.getgapup/mcp-knowledge","count":50,"changes":[{"tool":"job_result","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Poll the result of any tool called with async:true. Returns status=pending while running, status=completed with the full result once done, status=failed on error, or status=not_found if the job_id is unknown or expired (TTL 24h).","description_similarity":null,"before_hash":"14ac66f2ecb795bbca41c7b6f5c1988d","after_hash":"14ac66f2ecb795bbca41c7b6f5c1988d"},{"tool":"webhooks_manage","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Manage HTTP webhook callbacks for async tools (T5/T6 batch flagships). Instead of polling every 5s, register a callback URL — Gapup posts the job result to your endpoint the moment it completes. Supported events: job.completed | job.failed | monitoring.alert | quota.threshold. Modes: register (add endpoint), list (view active webhooks), revoke (soft-delete), test (fire a test payload to verify your receiver), history (last 20 fires). Security: every delivery is signed with HMAC-SHA256 on the body — verify the X-Gapup-Signature header against sha256(secret, body).","description_similarity":null,"before_hash":"b8be1a8968a3a61637d21a2ac514d29a","after_hash":"b8be1a8968a3a61637d21a2ac514d29a"},{"tool":"crm_connector","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Push, update, search and log activities in HubSpot, Salesforce or Pipedrive. 4 modes: push_lead (create contact/lead), update_opportunity (update deal stage/amount), search_contact (lookup by email), log_activity (call/email/meeting/note). Returns resource_id, direct CRM URL, signals and quality_score. If credentials are absent, returns a mock result with a warning signal. Auth: HubSpot via Bearer access_token; Salesforce via access_token + base_url; Pipedrive via api_key.","description_similarity":null,"before_hash":"35be64e5f66bbf32830e7dbc19449c57","after_hash":"35be64e5f66bbf32830e7dbc19449c57"},{"tool":"tool_recommend","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Cross-tool recommendation system: given a free-text intent, returns the most appropriate tools from the 170+ Gapup MCP catalogue, ranked by confidence, with pre-filled input suggestions and an optimal multi-tool chain when applicable. Use this first when you are unsure which tool to call — it navigates the full catalogue for you. Supports 15+ static pre-designed chains for frequent intents (M&A due diligence, sanctions screening, ESG 360, AI Act compliance, FTO patent clearance, crypto wallet tracking, etc.). Domains: compliance | finance | intel | legal | content | data | trade | infra. Pure compute — $0.01/call, no external fetch. Ideal as a first call in any multi-step agent workflow.","description_similarity":null,"before_hash":"24a33945688d56433f5a55509ed23bd3","after_hash":"24a33945688d56433f5a55509ed23bd3"},{"tool":"competitor_pricing_scrape","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Scrape and parse a competitor pricing page from a URL or domain. Fetches via proxy-aware timedFetch (tries /pricing, /plans, homepage fallback), then extracts: plan names, prices, billing cadence (monthly/annual/usage-based/one-time), key features, free tier presence, enterprise tier, estimated price range. Returns structured pricing tiers. If unfetchable or no pricing found (anti-bot, SPA, auth wall): returns a clear degraded result with warnings and signals — never fake success. ICP: founders, product managers, pricing strategists, competitive intel teams. Proxy-aware (AICI_RESEARCH_PROXY_URL). Cache TTL 6h.","description_similarity":null,"before_hash":"7ac36c8f80933297ae3210e786cfb52d","after_hash":"7ac36c8f80933297ae3210e786cfb52d"},{"tool":"seo_keyword_research","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"SEO keyword research from a seed keyword or topic. Uses Google Suggest (public, keyless) to discover related queries at 2 expansion levels, then clusters them by intent: informational / commercial / transactional / navigational — via heuristic pattern matching. Search volume is bucketed (very_high / high / medium / low / very_low) and clearly labelled as ESTIMATED — no fabricated precise numbers. Returns all keywords, intent clusters, quality scores (0-100), and top 10 opportunities. Supports country (gl) and language (hl) targeting. 100% keyless. Cache TTL 6h. ICP: SEO managers, content strategists, SaaS founders, agency teams.","description_similarity":null,"before_hash":"a5e0d75270f7584884107539e31f2ff3","after_hash":"a5e0d75270f7584884107539e31f2ff3"},{"tool":"seo_cro_audit","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Full SEO + CRO audit of any public URL. Analyses technical SEO (HTTP status, HTTPS, title/meta/canonical/robots, H1-H2, JSON-LD structured data, sitemap, robots.txt, OG/Twitter cards), content SEO (word count, keyword density top-10, readability estimate, image alt coverage, internal/external links), performance signals (page size, estimated render time, inline scripts/styles, unoptimised images), and CRO (CTA detection, above-fold CTAs, forms, social proof, trust signals, pricing visibility). Optionally compares up to 5 competitor URLs. Returns 0-100 scores per dimension plus a prioritised (P0/P1/P2) recommendation list. ICP: marketing managers, SEO/CRO consultants, e-commerce ops, agency teams. Budget: 8s per URL. Cache TTL: 1h.","description_similarity":null,"before_hash":"5727605fdde3d7338b8c0a86d2436d2d","after_hash":"5727605fdde3d7338b8c0a86d2436d2d"},{"tool":"transcribe_chapterize_media","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Transcription and chapterization of long-form media (YouTube, podcasts, direct audio/video) for content marketing teams, podcast publishers, edu tech, journalists and accessibility/compliance.\n\nPipeline:\n• YouTube → timedtext captions (keyless) + oEmbed metadata + native timecode chapters from description\n• Podcast RSS → episode description + duration + timecodes if embedded in show notes\n• Direct media → partial (requires Whisper API via OPENAI_API_KEY + force_whisper:true)\n• Chapters: native YouTube timecodes preferred; heuristic TF-IDF segmentation as fallback\n• Summary: extractive TF-IDF top-sentences (no LLM required)\n• Language detection: character-set heuristic (CJK→zh, kana→ja, hangul→ko, accents→fr/de/es)\n\nOutput formats: json (full structured object) | text (plain transcript) | srt | vtt\n\nSLA: ≤15s budget total. Cache: 24h TTL.","description_similarity":null,"before_hash":"d1e0dc9f5062353fac14e18ee931a901","after_hash":"d1e0dc9f5062353fac14e18ee931a901"},{"tool":"legal_clause_extractor","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Structured extraction of clauses, obligations and deadlines from legal documents (SaaS contracts, NDAs, employment agreements, loan agreements, leases, M&A deals, IP licences). Complements contract_risk_scanner with granular per-clause output.\n\nICP: legal ops, M&A lawyers, paralegals, contract managers, compliance officers.\n\nCapabilities:\n• Auto-detects document type (7 types) and language (EN/FR/DE/ES/PT)\n• Extracts parties with roles (buyer, seller, licensor, employee, etc.)\n• Splits document into sections and classifies 16+ clause types\n• Per-clause: 20 obligation patterns (EN/FR/DE), 10 deadline patterns, 18 risk detectors\n• Document-level: red flags (liability cap, auto-renewal, IP overreach, etc.), missing clauses per doc type\n• Global deadline calendar with P0/P1/P2 severity\n• Cross-reference map between sections\n• Cache: 7 days (legal docs stable once provided)\n\n100% pure compute — no external fetch required. Accepts 10k–100k char documents.","description_similarity":null,"before_hash":"1ea40c77413f31f8ce1710a80e871661","after_hash":"1ea40c77413f31f8ce1710a80e871661"},{"tool":"candidate_screening_ranking","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"AI-powered candidate screening and ranking for recruiters, hiring managers, ATS providers and recruitment AI agents. Ingests a job description and 1-50 candidate resumes, returning a ranked shortlist with score breakdowns across five weighted criteria: skills_match (tech stack and soft skills extracted from JD vs resume), experience_match (years vs seniority level inferred from JD), education_match (degree level + top-school detection), role_progression (Junior to Senior to Lead patterns), culture_fit_estimate (remote/hybrid, startup vs enterprise). Per candidate: overall_score 0-100, matched/missing skills, red_flags (job hopping, employment gaps, seniority mismatch), green_flags (long tenure, promotions), 3-5 interview questions, fit_summary. Diversity signals are first-name proxies ONLY with mandatory ethical WARNING. All processing is local -- no external API calls, instant response, privacy-preserving.","description_similarity":null,"before_hash":"0225621701786053a5a88c6998d009dd","after_hash":"0225621701786053a5a88c6998d009dd"},{"tool":"historical_price_series","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Fetch historical OHLCV price series for any ticker: stocks (AAPL, SAP.DE, 7203.T), ETFs, indices, commodities (GC=F for gold) or cryptocurrencies (BTC-USD). Returns a full date-indexed series of open/high/low/close/volume plus pre-computed statistics: total return, annualised return (CAGR), annualised volatility, max drawdown and Sharpe estimate (rf=4%). Automatically detects crypto tickers (→ CoinGecko) vs traditional assets (→ Yahoo Finance primary, Stooq fallback). Adjusts for dividends and splits when adjusted=true (default). Use cases: backtesting, factor analysis, performance attribution, charting, financial modelling. Sources: Yahoo Finance, CoinGecko, Stooq. All keyless. Optional env: AICI_RESEARCH_PROXY_URL for Bright Data routing (lifts Yahoo 429), TWELVE_DATA_API_KEY for higher Twelve Data quota.","description_similarity":null,"before_hash":"5d1296813fb6e80b91b0a078686b6c7b","after_hash":"5d1296813fb6e80b91b0a078686b6c7b"},{"tool":"workflow_orchestrator","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Meta-tool that CHAINS multiple MCP tools sequentially into a named workflow — delivering a composite output in a single call. 10 predefined workflows: compliance_full_audit (6 steps: KYC+sanctions+AI_gov+privacy+ESRS+CSRD), deal_due_diligence (7 steps: deep_dive+registry+court+patents+KYC+financials+M&A), market_entry_brief (6 steps: country_study+regulations+procurement+tax+AGOA+market_brief), competitor_intelligence_pack (5 steps: deep_dive+intel+patents+earnings+pitch_deck), esg_360 (5 steps: ESG_audit+carbon+CSRD+ESRS+supplier_esg), ip_freedom_to_operate (4 steps: patent_search+async_deep+IP_audit+competitive), climate_property_assessment (3 steps: climate_risk+real_estate+geo), pharma_target_screen (4 steps: trials+adverse_events+patents+meta_analysis), sanctions_360 (5 steps: KYC+Russian_sec+registry+crypto_wallet+court_filings), talent_market_brief (4 steps: salary+trends+adjacent_roles+skills_taxonomy). Returns steps_executed, consolidated P0/P1/P2 signals, overall_status, estimated_cost_usd, and raw outputs per step. Cache: 1h LRU per (workflow, target). Budget: 60s global timeout → partial if exceeded. Use when an agent needs a composite liverable without orchestrating tools manually.","description_similarity":null,"before_hash":"ac72564268e254c9c79697a80336cc19","after_hash":"ac72564268e254c9c79697a80336cc19"},{"tool":"job_postings_intelligence","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Agrégation d'offres d'emploi publiques pour inférer les tendances de recrutement. Trois modes : (1) company_hiring — analyse des postings d'une société : volume, fonctions (engineering/sales/marketing/ops/finance/hr), seniorité, géographie, croissance vs période précédente, signaux stratégiques inférés ; (2) role_market — volume marché global pour un rôle (open positions estimate, top employeurs, compétences demandées, médiane seniorité) ; (3) competitor_hiring_comparison — comparaison multi-sociétés (total postings, growth%, focus areas). Sources : Adzuna (ADZUNA_APP_ID/KEY env), RemoteOK (keyless), Himalayas (keyless), baseline statique 40 top employeurs. Usages : due diligence VC, intelligence compétitive, benchmarks RH, signaux pivots stratégiques. Cache 6h. SLA ≤15s.","description_similarity":null,"before_hash":"1e0923c1ae57c5aa981cfeb4308dccc8","after_hash":"1e0923c1ae57c5aa981cfeb4308dccc8"},{"tool":"climate_scenario_rcp","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Projections climatiques long terme par scénario IPCC (RCP AR5 + SSP AR6) pour toute localisation. Scénarios : RCP_4_5, RCP_8_5 (AR5), SSP1_2_6, SSP2_4_5, SSP3_7_0, SSP5_8_5 (AR6), ou 'all' (compare tous). Horizons : 2030–2100. Métriques : température (delta vs baseline 1990-2010, jours >35°C, nuits chaudes), précipitations (delta%, événements extrêmes, sécheresses), hausse du niveau de la mer (cm vs 2000), événements extrêmes (ouragans, inondations P100, sécheresses), indice incendie. Sorties : comparaison multi-scénarios, probabilité IPCC, signaux d'impact business par secteur. Sources : Open-Meteo CMIP6 (keyless), IPCC AR6 Atlas lookup, NOAA SLR projections. Usages : TCFD/CSRD physical risk, due diligence actifs long terme, assurance catastrophe, planification infrastructure. Cache 7j. SLA ≤20s.","description_similarity":null,"before_hash":"59dc6f75fac13118e12ca7eadeb2788f","after_hash":"59dc6f75fac13118e12ca7eadeb2788f"},{"tool":"china_ecommerce_intel","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Chinese e-commerce intelligence for the ZH diaspora (50M+), import-export teams, brand IP enforcement, MENA/Africa entrepreneurs sourcing from China, and brand monitoring. Covers Taobao, Tmall, JD.com, Pinduoduo, 1688.com (B2B) and AliExpress (cross-border).\n\nFive modes:\n• product_search — search products by keyword across CN platforms. Returns title ZH/EN, price CNY + USD estimate, sales 30d, rating, seller info, product URL.\n• seller_profile — full seller/supplier dossier: factory vs reseller detection, certifications (ISO, BSCI, CE), rating, years in business, main categories.\n• price_history — 12-month price trend for a product (live current price + seasonal model for CN shopping festivals: 11.11, 6.18, CNY).\n• brand_monitoring — detect counterfeits and grey market listings: price anomaly detection (>50% below MSRP = suspicious), counterfeit keyword scan, risk score 0-100.\n• market_intel — category overview: top 5 sellers by market share, avg/median price, volume estimate, price range.\n\nData quality note: LIVE data from Taobao/Tmall/JD/Pinduoduo REQUIRES AICI_RESEARCH_PROXY_URL with CN residential routing (Bright Data -country-cn). Without proxy: AliExpress (cross-border) + curated category fallback available.\n\nInput formats for seller_profile: 'platform:id' e.g. 'aliexpress:123456', '1688:87654321', 'tmall:apple-store-official'.\nInput formats for price_history: AliExpress product URL or numeric product ID.","description_similarity":null,"before_hash":"f55026a8b0c00639e719960630431247","after_hash":"f55026a8b0c00639e719960630431247"},{"tool":"ugc_moderation_classifier","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Multi-language UGC content moderation for marketplaces, social platforms and comment systems. Detects policy violations in text content across 9 policies and 12 languages without external API calls.\n\nPolicies checked:\n• hate — hate speech, slurs, dehumanization (50+ terms × 12 languages)\n• sexual — explicit sexual content, pornography references, nudity solicitation\n• violence — threats, weapon references, graphic violence\n• self_harm — suicidal ideation, self-injury, eating disorder promotion\n• harassment — doxxing, stalking, cyberbullying, blackmail\n• scam — phishing, investment fraud, romance scam, lottery fraud\n• spam — bots, keyword stuffing, excessive caps, emoji storms, suspicious URLs\n• copyright — piracy, leaked content, serial keys, streaming fraud\n• minor_safety — grooming signals, CSAM references, minor + adult content combos\n\nLanguages: en / fr / de / es / it / pt / nl / zh / ja / ko / ar / ru (auto-detected)\n\nOutput includes severity (low/medium/high/severe), confidence (0-100), matched patterns, excerpt, recommended action, age appropriateness (adult/teen/child), and signals.\n\nNo API key required. Stateless — no content is stored or logged.","description_similarity":null,"before_hash":"b0d3882a63d55c7ef92618e8965f3b19","after_hash":"b0d3882a63d55c7ef92618e8965f3b19"},{"tool":"arbitration_awards_lookup","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Commercial arbitration intelligence for litigation lawyers, M&A due diligence teams, sovereign wealth funds and trade finance compliance. Covers 8 major institutions: ICC, AAA, LCIA, HKIAC, SIAC, CIETAC, DIAC, ICDR.\n\nThree modes:\n• party_lookup — find awards by party name (searches 20 landmark public awards + JusMundi best-effort)\n• institution_index — browse awards and caseload stats per institution with date range filter\n• clause_check — audit an arbitration clause for missing elements (institution, seat, language, arbitrator count, governing law, binding nature)\n\nNote: Most arbitration awards are confidential. This tool surfaces public awards (Yukos, Crystallex, Achmea, etc.) plus redacted statistics from institutional annual reports. Private awards are not accessible.\n\nCache: 24h (arbitration data is very stable). No API key required.","description_similarity":null,"before_hash":"3bd569231092844c2c95db974edadcc7","after_hash":"3bd569231092844c2c95db974edadcc7"},{"tool":"earnings_transcript_signals","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Earnings call transcript signal extractor for equity research analysts, catalyst-driven hedge funds, and BD teams. Parses earnings transcripts (fetched or provided) to surface:\n\n• signals (P0/P1/P2): guidance raise/cut, miss/beat vs consensus, buyback, dividend change, new product, executive change, capex shift, M&A intent, regulatory risk, competitive threat, supply chain, hiring\n• kpis_mentioned: Revenue, EBITDA, EPS, FCF, Gross Margin, Operating Margin with YoY/QoQ %\n• guidance: raised / maintained / cut / new_initiated items extracted\n• q_and_a_topics: top Q&A themes detected (AI strategy, China exposure, M&A pipeline, macro, etc.)\n• overall_tone: bullish / neutral / bearish\n\nSources fetched automatically: SEC EDGAR 8-K filings, Yahoo Finance earnings news, Motley Fool transcripts. If no transcript can be retrieved from any source, returns status:'failed' with an explicit warning and empty signals — never fabricated data. Accepts transcript_text override for direct analysis. Supports multilingual transcripts (de/fr/es/zh). European tickers (SAP.DE, BMW.DE) mapped to EDGAR-compatible equivalents automatically.","description_similarity":null,"before_hash":"f75772d9dfc5b2eff0721cc4eb18492d","after_hash":"f75772d9dfc5b2eff0721cc4eb18492d"},{"tool":"monte_carlo_portfolio","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Pure-compute Monte Carlo portfolio simulation using Geometric Brownian Motion (GBM). Models a multi-asset portfolio across time with contributions, withdrawals, and annual rebalancing. Returns full probability distribution of terminal wealth, percentile paths, drawdown stats, and Sharpe ratio. Modes: simulate (full Monte Carlo) | glide_path (lifecycle 110-age target-date allocation) | stress_test (4 historical crises: 2008 GFC / 2000 dotcom / 1970s stagflation / 2020 COVID). No external data needed — all computed from asset assumptions. Ticker defaults built-in: SPY/VOO/VTI 7%/15%, QQQ 9%/20%, TLT/BND 3%/6%, GLD 5%/18%, BTC 30%/70%. ICP: asset managers, family offices, retail wealth advisors, robo-advisor agents, retirement planners. 10k simulations × 30 years runs in <3s on V8 JIT.","description_similarity":null,"before_hash":"1f9910c4cd69dd657f1d562b12274fe0","after_hash":"1f9910c4cd69dd657f1d562b12274fe0"},{"tool":"crypto_wallet_intel","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Multi-chain on-chain analytics for crypto trading agents, on-chain analysts, AML/compliance teams and DeFi BD. Covers Ethereum, Base, Polygon, BSC, Arbitrum, Optimism — EVM-compatible addresses only.\n\n5 modes:\n• wallet_profile — full wallet summary: type (EOA/contract/CEX/protocol), inferred persona (whale/MEV-bot/DeFi-user/hodler…), age, tx count, native balance, ERC-20 count, NFT collections, OFAC sanctions flag\n• token_flows — ERC-20 inflows/outflows per token on the selected period, priced in USD via CoinGecko\n• pnl_estimate — FIFO realized + unrealized P&L on the period with confidence rating (high/medium/low)\n• counterparties — top 20 counterparties ranked by USD volume with CEX/DEX/protocol labels\n• defi_positions — active DeFi positions detected via Etherscan interaction history (Aave/Compound/Uniswap/Curve/Lido/Balancer/SushiSwap)\n\nSignal detection (P0/P1/P2):\n  P0 if OFAC SDN match OR direct Tornado Cash / sanctioned-protocol interaction\n  P1 if >$1M volume on wallet <30 days old OR MEV-bot pattern OR >80% volume on single counterparty\n  P2 informational (CEX wallet, new wallet, no anomaly)\n\nSources: Etherscan family (keyless free-tier, optional API key per chain), DefiLlama (keyless), public EVM RPC (keyless), CoinGecko free tier (keyless).\nCache TTL: 5 min (wallet activity evolves fast). Budget: 8s per source.\n\nEnv vars (all optional, raise Etherscan rate-limit from 1 req/5s to 5 req/s):\n  ETHERSCAN_API_KEY · BASESCAN_API_KEY · POLYGONSCAN_API_KEY\n  BSCSCAN_API_KEY · ARBISCAN_API_KEY · OPTIMISM_API_KEY","description_similarity":null,"before_hash":"2df27972d0d08baf370bac38de934d9a","after_hash":"2df27972d0d08baf370bac38de934d9a"},{"tool":"financial_model_3statement","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Pure-compute 3-statement financial model builder (Income Statement + Balance Sheet + Cash Flow). Feed assumptions (revenue growth, COGS%, OpEx, CapEx, working capital, tax rate, depreciation, debt schedule) → receive a full 3-5 year projection with integrated DCF valuation. Supports IFRS / US_GAAP / PRC_GAAP (中国会计准则) norms with bilingual ZH+EN labels for PRC. Modes: build (full 3-statement model) | scenario_analysis (base/bull/bear ±20% growth) | sensitivity (1 KPI × 1 input, 5-point grid). No external data needed — all computed from assumptions. ICP: VC due diligence, M&A analysts, CFO SMB, startup founders pitching investors, biotech/SaaS modeling. Returns balance_check_ok per year, DCF enterprise/equity value, and coherence warnings.","description_similarity":null,"before_hash":"135a7584268b7dff5de1207acbca3478","after_hash":"135a7584268b7dff5de1207acbca3478"},{"tool":"esg_audit_multi","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Multi-mode ESG intelligence for ESG analysts, sustainability officers and impact investing fund managers. Aggregates live data from CDP, SBTi, Wikipedia, Yahoo Finance and web search across five modes:\n• company_score — ESG score 0-100 with E/S/G breakdown + heuristic rating (AAA-CCC), from CDP grade + SBTi + sector profile\n• controversy_check — controversies detected via web search, classified P0/P1/P2 by type (greenwashing, emissions fraud, labour, governance)\n• emissions — GHG Scope 1/2/3 estimates, SBTi validation flag, net-zero target year, carbon intensity per M€ revenue\n• esrs_readiness — CSRD gap across 12 standards (E1-E5, S1-S4, G1-G3): readiness % + gap list + CSRD deadline + effort man-days\n• sfdr_classification — suggested SFDR Article 6/8/9 with rationale and sustainability indicators met\n\nSignals: P0=critical (controversy/score<40), P1=significant (score<55/SBTi missing/ESRS<50%), P2=watch. Cache 24h.","description_similarity":null,"before_hash":"00ee481ae779c2103947bf1354508fe4","after_hash":"00ee481ae779c2103947bf1354508fe4"},{"tool":"agoa_eba_intelligence","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Intelligence préférentielle AGOA (US→Africa) et EBA/GSP (EU→Africa). Vérifie l'éligibilité d'un pays africain aux programmes tarifaires préférentiels, l'éligibilité d'un produit par code HS, identifie les meilleures opportunités d'export Afrique→US/EU, et fournit les règles de conformité (rules of origin, valeur ajoutée, docs). Différenciateur Africa diaspora : 39 pays AGOA + 47 LDCs EBA encodés. Sources : AGOA.info · EU EBA · EU GSP+ · WTO Tariff · UN Comtrade.","description_similarity":null,"before_hash":"7995b07950e9114ef1300db69c0738dd","after_hash":"7995b07950e9114ef1300db69c0738dd"},{"tool":"talent_intelligence","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"HR tech intelligence for CHROs, recruiters, VC teams, comp & benefits leads and workforce planners. Four modes powered by ESCO, O*NET, BLS OES and crowd-sourced salary data:\n\n• salary_benchmark — cash-only salary medians (p25/median/p75) for 54+ roles across US/EU/Asia. Covers tech, finance, compliance, healthcare, marketing, ops and C-suite. Data from BLS OES, Levels.fyi and StackOverflow Developer Survey 2024.\n• skills_taxonomy — maps a skill to its ESCO URI, O*NET codes, skill type (hard/soft/knowledge/cert), 8 related skills with similarity scores and typical roles.\n• job_market_trends — YoY growth %, open positions estimate, top employers and leading skills per job category × country. Static 2024 data with BLS baseline fallback.\n• adjacent_roles — up to 6 roles adjacent to a source role with ESCO taxonomy adjacency: similarity score, salary delta % and skills overlap %.\n\nAll salary data is cash-only (excludes equity/RSU/bonus). Cache TTL: 24h (stable labour market data). Optional env ONET_API_KEY for authenticated O*NET lookups (free registration at onetcenter.org).","description_similarity":null,"before_hash":"0a6c33726e144d76dc94354ae380de33","after_hash":"0a6c33726e144d76dc94354ae380de33"},{"tool":"tax_compliance_multi","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Multi-jurisdiction tax compliance data for international SaaS, cross-border marketplaces and expat services. Five modes: (1) vat_lookup — validate EU VAT numbers live via VIES SOAP (27 EU countries) or UK VRN via HMRC; (2) sales_tax — US state sales tax rates, nexus thresholds (post-Wayfair 2018), digital goods taxability for all 50 states + DC; (3) gst — APAC GST/SST/consumption-tax rates for IN, SG, AU, NZ, MY, JP, KR, TH, ID, PH, VN with reduced rates and registration thresholds; (4) oss_ioss_eligibility — EU One-Stop-Shop and Import-OSS eligibility analysis (EUR 10k OSS threshold, EUR 150 IOSS per-consignment); (5) transfer_pricing_benchmark — OECD/JTPF operating-margin benchmarks by industry and country (20+ sectors, country-specific adjustments). Returns P0/P1/P2 compliance signals: P0=invalid VAT used for zero-rating, P1=taxable digital goods detected/audit risk, P2=filing deadlines/nexus alerts. Keyless — no API key required. Optional env: HMRC_VAT_API_KEY for UK VAT live validation. Cache TTL 24h.","description_similarity":null,"before_hash":"8e2dafd847813d68da22a6ad3f2c5e39","after_hash":"8e2dafd847813d68da22a6ad3f2c5e39"},{"tool":"geo_logistics_intel","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Geospatial logistics intelligence for supply chain, maritime and transport agents. Four modes: (1) geocode_batch — resolve up to 50 addresses to lat/lon with confidence scores (OSM Nominatim + Open-Meteo fallback, 1 req/s rate-limit respected); (2) routing — road/cycling/walking route with distance_km, duration_seconds and ETA ISO timestamp between two addresses or lat/lon points (OSRM public, keyless, global); (3) port_congestion — congestion status for any UN/LOCODE port (e.g. NLRTM, SGSIN, CNSHA) with waiting vessel count, severity (low/medium/high/extreme) and average wait hours; (4) ship_tracking — AIS position, speed, course, destination and ETA for a vessel by its 9-digit MMSI. No API key required for geocode/routing/port. Optional env: AIS_STREAM_API_KEY for live ship data (otherwise MarineTraffic scrape best-effort). SLA: <=25s p95. Cache: 24h geocoding / 1h routing / 30min port / 5min ship. Quality score 0-100. Status: final/partial/failed.","description_similarity":null,"before_hash":"109ab0aca996b31163f7616f1843e031","after_hash":"109ab0aca996b31163f7616f1843e031"},{"tool":"sharia_compliance_screener","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Sharia compliance screening engine for Islamic banks, Sukuk issuers, Gulf sovereign funds, halal investment managers and MENA family offices. Zero competing MCP on this vertical.\n\nStandards supported: AAOIFI (default) | MSCI_Islamic | S&P_Sharia | DJIM\n\nFour modes:\n• company — Full Sharia screen of a listed company: business activity (halal/haram/mixed) + AAOIFI financial ratios (debt/market-cap <30%, interest-assets <30%, non-compliant revenue <5%)\n• instrument — Sukuk / halal fund classification by ISIN or name. Maps to known Sharia boards.\n• sector_screen — Industry classification (halal/haram/mixed) with rationale + examples. Static AAOIFI-based map covering 40+ sectors.\n• financial_ratios — AAOIFI ratio computation on fetched or provided financials.\n\nProhibited activities screened: alcohol, gambling, pork, weapons, pornography, tobacco, conventional banking (riba), conventional insurance, adult entertainment, embryonic stem cells.\n\nOutput includes compliance_status (halal/haram/doubtful_mixed/purification_required), purification_pct when applicable, P0/P1/P2 signals, quality_score, and sources.","description_similarity":null,"before_hash":"f0bf5bbf30206acf7997d53053cd99a1","after_hash":"f0bf5bbf30206acf7997d53053cd99a1"},{"tool":"email_domain_health_check","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Comprehensive email domain health check: MX routing, SPF authentication, DKIM signing, DMARC policy enforcement, DNSBL blacklist status (Spamhaus/SpamCop/Barracuda), TLS certificate validity, and WHOIS registration age. Aggregates a reputation score 0-100 and generates P0/P1/P2 deliverability signals. Accepts a domain (stripe.com) or email address (info@stripe.com). Detects role-based addresses (info@, support@, admin@, noreply@) that have higher bounce rates. Detects email provider (Google Workspace, Microsoft 365, Amazon SES, etc.). P0 signals: blacklisted / no MX / TLS expired / no SPF + DMARC none. P1 signals: SPF soft-fail / no DKIM selector / DMARC no reporting. P2 signals: role-based address / TLS expires <30d / domain age <90 days. All checks are keyless (no API keys required). Cache TTL 1h. SLA <=10s p95.","description_similarity":null,"before_hash":"f9e308260ed89b1773628134a4ff579e","after_hash":"f9e308260ed89b1773628134a4ff579e"},{"tool":"india_market_data","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Indian capital market intelligence for the IN diaspora (30M+) and investors. Covers NSE, BSE, and MCA corporate registry across four modes:\n\n• company — full company profile: name, CIN, exchange, NSE/BSE tickers, industry, incorporation date, paid-up capital, registered office, status, directors\n• market_quote — real-time quote: price (INR), change%, volume, market cap, P/E ratio. Sources: Yahoo Finance (primary), BSE API, NSE API (proxy-gated)\n• sector_overview — Nifty/Sensex sector snapshot: top 5 companies by market cap. Supported sectors: it, banking, pharma, energy, auto, fmcg, realestate, metals, telecom, consumer\n• mca_filing — Ministry of Corporate Affairs filings. Requires CIN for direct lookup.\n\nInput formats accepted:\n  • NSE ticker (e.g. 'RELIANCE', 'TCS.NS')\n  • BSE 6-digit code (e.g. '500325' for Reliance)\n  • CIN 21-char (e.g. 'L17110MH1973PLC019786')\n  • Company name EN (e.g. 'Reliance Industries', 'Tata Consultancy')\n  • Sector keyword (e.g. 'IT services', 'banking', 'pharma')\n\nENV: AICI_RESEARCH_PROXY_URL with country-in routing unlocks NSE direct API and MCA.","description_similarity":null,"before_hash":"32a2e2851c6d5706f3485a6c5ce5a6e7","after_hash":"32a2e2851c6d5706f3485a6c5ce5a6e7"},{"tool":"china_market_data","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Chinese capital market intelligence for the ZH diaspora (50M+) and institutional investors. Covers A-Shares (SSE/SZSE), H-Shares (HKEX), and ADRs across four modes:\n\n• company — full company profile: name ZH/EN, USCC (18-digit social credit code), exchange, industry (CSRC classification), chairperson, registered capital, SOE flag\n• market_quote — real-time quote: price (CNY or HKD), change%, volume, market cap, P/E ratio, dividend yield, last update timestamp\n• sector_overview — sector snapshot: top 5 companies by market cap, avg P/E, 30-day sector index change. Supported sectors: semiconductor, ev, battery, technology, finance, energy, realestate, consumer, pharma, telecom\n• regulatory_filing — recent regulatory disclosures (HKEX filings: annual, quarterly, announcements, mergers, IPOs) with title, date, document URL\n\nInput formats accepted:\n  • 6-digit A-Share ticker (e.g. '600519' for Moutai SSE)\n  • HKEX ticker (e.g. '0700.HK' or '700' for Tencent)\n  • Company name in EN or ZH (e.g. '腾讯', 'Kweichow Moutai')\n  • Sector keyword (e.g. 'semiconductor', '半导体')\n\nData sources: Yahoo Finance (primary, always accessible), Eastmoney push2 + CompanySurvey (via Bright Data proxy when AICI_RESEARCH_PROXY_URL is set), HKEX filing API. Note: Eastmoney/CSRC/SSE are blocked from datacenter IPs without proxy — set AICI_RESEARCH_PROXY_URL to unlock full coverage.","description_similarity":null,"before_hash":"64d60fad2bef8daf31272a764296a1d8","after_hash":"64d60fad2bef8daf31272a764296a1d8"},{"tool":"patent_landscape_result","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Poll the result of a patent_landscape_async job. Returns status=pending while running, status=completed with the full patent landscape report once done, status=failed on error, or status=not_found if the job_id is unknown or expired (TTL 24h). Call this after the eta_seconds hint returned by patent_landscape_async (~180s).","description_similarity":null,"before_hash":"a5e275d9ffcf5b5d6e42ee6b1fac0b07","after_hash":"a5e275d9ffcf5b5d6e42ee6b1fac0b07"},{"tool":"patent_landscape_async","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Async extended variant of patent_landscape. Supports max_results up to 200 (vs 50 in sync mode) and an optional include_citation_graph flag that enriches each patent with its 2-level citation graph (parent patents that cite this one + child patents cited by this one). Returns immediately (<300ms) with a job_id. Poll the result with patent_landscape_result(job_id) after eta_seconds (~180s). Use for deep R&D white-space analysis, freedom-to-operate (FTO) audits, VC due diligence IP mapping, or large-scale competitor portfolio analysis. Async tool — register a webhook via `webhooks_manage(register, url, [job.completed])` to receive callbacks instead of polling. Faster + lighter.","description_similarity":null,"before_hash":"69dd8d8df8ff7abdeaba41ffc4415a86","after_hash":"69dd8d8df8ff7abdeaba41ffc4415a86"},{"tool":"sci_literature_search","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Recherche bibliographique multi-sources sur la litterature scientifique. Sources : OpenAlex (200M+ works) · Semantic Scholar · arXiv · PubMed · CrossRef. Modes : search | meta_analysis | citation_network | expert_finder. Keyless / free tier. Cache LRU 12h.","description_similarity":null,"before_hash":"338945efba56298ea5774139d398e799","after_hash":"338945efba56298ea5774139d398e799"},{"tool":"usdc_x402_payments_intel","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Real-time analytics on x402 protocol USDC micropayments for MCP endpoints on Base network. Unique competitive advantage: aggregates internal production telemetry (our own traffic data) with on-chain USDC Transfer events and Bazaar marketplace listings — data no external competitor can access. Four modes: (1) facilitator_stats — Coinbase x402 facilitator settlement statistics (volume, count, top payees/payers). Uses Coinbase CDP API if COINBASE_X402_API_KEY is set; falls back to Base mainnet RPC scan of USDC transfers to known facilitator addresses. (2) endpoint_intel — Per-MCP-endpoint analytics: tx count, USDC volume, unique callers, success rate, catalog size. For gapup-mcp.io endpoints: reads internal JSONL telemetry (richest data source, unique). (3) agent_caller_profile — Anonymous profile of a calling agent wallet: tx count, USDC spent, top endpoints, inferred persona (depth-seeker / bulk-scanner / generalist / researcher / explorer). Wallet anonymised via SHA-256. (4) price_radar — USDC price distribution by tool category (data_lookup / synthesis / compliance / competitive) from Bazaar + internal catalog. Returns median, P25, P75. Network: Base mainnet. USDC contract: 0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913. Cache: 30 min LRU. Timeout per source: 8s. Optional env: COINBASE_X402_API_KEY (higher-fidelity facilitator stats).","description_similarity":null,"before_hash":"45464a368bdab72f3251612cbcfeedd9","after_hash":"45464a368bdab72f3251612cbcfeedd9"},{"tool":"clinical_pharma_intel","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Clinical and pharmaceutical intelligence for biotech analysts, healthcare fund managers, pharma BD teams, catalyst-driven hedge funds and health journalists. Aggregates live data across five modes:\n• trials — active/completed clinical trials (ClinicalTrials.gov v2 + EU CTR in parallel, 450k+ records)\n• pipeline — full pipeline by sponsor: trial count by phase + top indications\n• approvals — FDA drug label approvals + mechanism of action (OpenFDA)\n• recalls — FDA enforcement recalls classified by severity (Class I/II/III)\n• adverse_events — FAERS aggregated reactions: top 10 reactions + serious% \n\nSignal detection (P0/P1/P2):\n  P0 if Class I recall OR trial terminated for safety reason\n  P1 if serious adverse events >30% OR ≥3 recalls in 12 months\n  P2 otherwise (standard monitoring)\n\nAll sources are public and keyless. Optional env OPENFDA_API_KEY raises daily quota from 1,000 to 120,000 requests. SLA: ≤16s p95 (parallel fetch, 8s budget per source). Cache: 6h trials, 24h approvals, 12h recalls, 6h adverse events.","description_similarity":null,"before_hash":"500a05388ed88a43790f225610872e97","after_hash":"500a05388ed88a43790f225610872e97"},{"tool":"real_estate_intel","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Real estate intelligence aggregator with a best-in-class French dataset (DVF — Demandes de Valeurs Foncières — 100% of FR transactions since 2019, public, keyless) plus UK Land Registry Price Paid (all UK transactions 1995+). Four modes: (1) property — full transaction history for a specific address; (2) comparables — median/std price/m² within a radius (default 500m); (3) market — annual price series, YoY change, volume, trend by commune; (4) valuation — two-method estimate (comparables median + hedonic regression if n≥30) with confidence scoring (high/medium/low). All sources are free and require no API key. ICP: PropTech agents, REITs, fund managers, family offices, insurance. SLA: ≤25s p95 (sources fetched in parallel, 8s budget each). Cache: 24h TTL (DVF data is stable). Quality score: 30 pts DVF retrieved, 20 pts geocoding, 20 pts UK LR retrieved, 15 pts if comparables count ≥10, 15 pts if method quality achieved. Status: failed/<60/≥60 → failed/partial/final. No env vars required.","description_similarity":null,"before_hash":"7b4b0319bb9e22a79fe72197f10d2ad1","after_hash":"7b4b0319bb9e22a79fe72197f10d2ad1"},{"tool":"weather_climate_intel","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Physical climate intelligence for insurance underwriting, agritech, logistics, energy trading and ESG/climate risk disclosure. Three modes: (1) forecast — 14-day daily weather forecast with temperature, precipitation, wind and humidity; (2) historical — daily records and monthly aggregates for any date range since 1940, with anomaly detection (P90/P95 heat events, extreme precipitation days); (3) climate_risk — long-term physical risk scoring combining CMIP6 ensemble projections (2020-2050), altitude, FEMA flood zones (US) and historical baselines. Risk dimensions: flood, heat (days >35°C/year), drought (SPI), wildfire, sea-level. Overall score 0-100 (100 = severe). Location: city string or lat/lon coordinates. Sources: Open-Meteo (keyless, global, 1940→2050), Open-Elevation, FEMA NFHL (US), NOAA CDO (optional NOAA_API_KEY env var for US+global station data). SLA: ≤25s p95. Cache: 1h forecast / 24h historical / 7d climate_risk.","description_similarity":null,"before_hash":"0b50e9c4e4cf91515dfe5cdbec2b4f83","after_hash":"0b50e9c4e4cf91515dfe5cdbec2b4f83"},{"tool":"patent_landscape","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Search, analyze and map patent landscapes across major jurisdictions (US, EP, WO, CN, JP, KR). Three modes: (1) search — find patents by keywords, company name or inventor name; (2) landscape — aggregate distributions: top assignees, top inventors, CPC class breakdown, filings by year, citation leaders, white-space innovation opportunities; (3) lookup — retrieve a specific patent by number (e.g. US10000000B2, EP3456789A1, WO2023/123456). Primary source: WIPO PatentScope (WO PCT, keyless). Optional sources: USPTO PatentsView (US, env PATENTSVIEW_API_KEY), EPO OPS (EP/WO, env EPO_OPS_CONSUMER_KEY + EPO_OPS_CONSUMER_SECRET), Lens.org (global, env LENS_API_TOKEN). Use cases: freedom-to-operate (FTO) analysis, R&D gap identification, VC due diligence IP audit, competitor patent portfolio mapping, inventor network analysis. SLA: <=24s p95 (parallel fetches, 8s per source). Cache: 24h TTL (patent data stable). Quality score: 30 pts per retrieved source (max 90), +10 if >=10 patents, +10 bonus for landscape mode with non-empty top_assignees.","description_similarity":null,"before_hash":"0e1c816af96804e036e2e9d7ba54d364","after_hash":"0e1c816af96804e036e2e9d7ba54d364"},{"tool":"dependency_vulnerability_scan","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"SCA (Software Composition Analysis) — scans a project dependency manifest and returns known vulnerabilities for each dependency. Supports: package.json (npm), requirements.txt (Python), go.mod (Go), Cargo.toml (Rust), composer.json (PHP), Gemfile.lock (Ruby), CycloneDX SBOM JSON. PRIMARY source: OSV.dev (keyless, free, covers npm/PyPI/Go/crates.io/Packagist/RubyGems + GHSA advisories federated). CVSS enrichment: NVD NIST (when OSV lacks score). Exploitation flag: CISA KEV (known-exploited-vulnerabilities catalog). Returns per-vuln CVE/GHSA IDs, severity, CVSS score, fixed version, and actionable upgrade recommendations. Relevant for EU NIS2 supply chain risk obligations, DORA, SOC 2 vendor assessments. Cache TTL 6h. Parallel OSV queries (concurrency=10). SLA <=30s p95.","description_similarity":null,"before_hash":"a76662220f9e9a07f336ba505c32b92a","after_hash":"a76662220f9e9a07f336ba505c32b92a"},{"tool":"cve_security_lookup","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Look up CVE vulnerability data for enterprise security teams, DevSecOps and SOC analysts. Supports two modes: exact CVE ID lookup (e.g. 'CVE-2024-3094') or keyword search by product/vendor (e.g. 'openssl', 'Apache Tomcat'). Cross-references four authoritative keyless sources: NVD NIST (official CVE database, CVSS v3 scores, affected CPEs), CISA KEV (Known Exploited Vulnerabilities catalog — exploit_in_wild flag), EPSS FIRST (exploit probability 0-1), GitHub Security Advisories (ecosystem-specific: npm/pypi/maven). Returns structured vulnerability records with CVSS v3 scores, affected product version ranges, CWE weakness classification, references and exploitation status. Signals engine produces P0/P1/P2 alerts: P0=CVSS>=9 + active exploitation, P1=CVSS>=7 or EPSS>=70%, P2=CWE pattern clusters. Relevant for EU NIS2 and DORA supply chain risk obligations. Optional env: NVD_API_KEY (raises NVD rate-limit 5→50 req/30s), GITHUB_TOKEN (raises GHSA GraphQL rate-limit). Cache TTL 6h. SLA <=25s p95.","description_similarity":null,"before_hash":"966ce9b399b0582b633f57f119018b2a","after_hash":"966ce9b399b0582b633f57f119018b2a"},{"tool":"web_search_multilang","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Multi-language, multi-source web search that goes beyond Anglo-centric results. Supports 15 languages (fr/de/es/it/pt/nl/ja/zh/ko/ar/ru/sv/pl/tr/en) with automatic detection. Aggregates results from Mojeek (independent search engine, multilang) and Wikipedia (native multilang API), with DDG and HN as English-language complements. Returns deduplicated results ranked by cross-engine consensus. Use when you need non-English search results, when DDG fails, or for geographically-biased queries. Phase 2 #7 of the geo/lang expansion plan. Note: Brave/Bing/Searx are blocked from DO IPs — configure AICI_RESEARCH_PROXY_URL for residential proxy.","description_similarity":null,"before_hash":"4b10965dafa6342ab89a9031a8881f4c","after_hash":"4b10965dafa6342ab89a9031a8881f4c"},{"tool":"gov_procurement_multi","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Aggregate public procurement tenders (calls for tender / appels d'offres) from multiple government sources simultaneously: TED Europa v3 (27 EU countries, keyless API), BOAMP France (opendatasoft, keyless), UK Contracts Finder (OCDS standard, keyless), SAM.gov United States (requires SAM_GOV_API_KEY env var), and bund.de Germany (HTML scraping, partial). Returns structured tender records with buyer authority, EU CPV sector code, estimated contract value converted to EUR via live FX rates, submission deadlines, and direct notice URLs. Use when: a B2G agent needs to find government contract opportunities matching keywords across multiple jurisdictions; building a pipeline of public tenders for bid/no-bid qualification; monitoring a domain by CPV code; market sizing public sector spend. Key inputs: query (keywords), countries (ISO-2 array), cpv_codes (EU standard codes, e.g. 72000000=IT services, 45000000=construction, 79000000=business services), min_value_eur (filter), published_after (ISO date, defaults to 30 days ago). SLA: <=25s p95 (all sources fetched in parallel, 8s budget per source). Optional env var SAM_GOV_API_KEY enables US federal tenders (free key at api.sam.gov). Quality score: 25 pts if TED EU retrieved, 15 pts per other source retrieved (max 60), 10 pts if >= 10 tenders returned, 5 pts if aggregates computed. Status: failed < 30 / partial 30-59 / final >= 60.","description_similarity":null,"before_hash":"bb40a95747bb3b91a69103db79293ba4","after_hash":"bb40a95747bb3b91a69103db79293ba4"},{"tool":"court_filings_multi","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Aggregate court filings, judgments and litigation records for a company or individual across five major legal jurisdictions: US (CourtListener / PACER), UK (National Archives — EWHC/EWCA/UKSC/UKUT), EU (ECHR HUDOC — European Court of Human Rights), France (Légifrance / Cour de cassation) and Germany (BGH / BVerfG). Returns structured case records with type classification (civil/criminal/antitrust/bankruptcy/administrative/unknown), status (filed/pending/decided/appealed/unknown), parties extracted from case titles, opinion URLs and verbatim snippets. Cross-case pattern recognition produces severity-ranked signals (P0–P2) for criminal, antitrust, bankruptcy, regulatory, data-breach and IP categories. Use when: due diligence on a counterparty, vendor risk assessment, competitive intelligence (litigation history), regulatory exposure mapping. All sources are public and keyless. Optional env var COURTLISTENER_API_KEY raises US rate limits beyond the default 5 req/s anonymous tier. SLA: ≤25s p95 (all jurisdictions fetched in parallel, 8s budget per source). Quality score: 20 pts per jurisdiction with ≥1 case retrieved, +10 if signals detected, +5–10 if ≥2–3 distinct sources contributed.","description_similarity":null,"before_hash":"e599b6c74709af35fb072b379a7d6981","after_hash":"e599b6c74709af35fb072b379a7d6981"},{"tool":"corporate_registry_lookup","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Resolve legal information about a company from its national corporate registry. Returns a normalised, sourced company profile: legal status, registration number, directors, shareholders, recent filings, registered address, share capital, and a quality score (0–100). Coverage: France (INPI, keyless — full SIREN/SIRET with directors), 3M+ entities worldwide via GLEIF LEI (keyless, large companies), UK (Companies House, optional key), Netherlands (KvK, optional key), and OpenCorporates (token required since 2026). Sources are tried in cascade; quality_score increases with each source that succeeds. When to use: due-diligence, KYC screening, supplier verification, M&A research, or any workflow needing verified company identity and legal status. Optional env vars: COMPANIES_HOUSE_API_KEY (UK), KVK_API_KEY (NL), OPENCORPORATES_API_TOKEN (OpenCorporates token).","description_similarity":null,"before_hash":"aa61eefb2c321c4a1450415fac127eae","after_hash":"aa61eefb2c321c4a1450415fac127eae"},{"tool":"economic_indicator","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Return a precise macroeconomic indicator for a country — the exact figure for a market-sizing, finance or strategy workflow. Indicators: gdp_usd, gdp_per_capita, gdp_growth, inflation, unemployment, population. Source: World Bank. When to use: an agent's analysis needs an authoritative country-level economic figure. Inputs: country (ISO-2 or ISO-3 code) and indicator name.","description_similarity":null,"before_hash":"9a2e9f683c6634dfb0bbb7bc334ff2b9","after_hash":"9a2e9f683c6634dfb0bbb7bc334ff2b9"},{"tool":"interest_rate","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Return a precise reference interest rate — the exact figure an agent injects into a treasury, lending, valuation or trading model. Available rates: fed_funds, sofr, us_10y, us_2y, us_3m, ecb_main, euribor_3m. Source: FRED (Federal Reserve Bank of St. Louis). When to use: an agent's computation needs a current benchmark rate as a precise input.","description_similarity":null,"before_hash":"f98dad1b9ce85e965813c8f3e9ddd441","after_hash":"f98dad1b9ce85e965813c8f3e9ddd441"},{"tool":"realtime_data_streams","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"High-frequency real-time market data for trading agents, market-making bots and fintech analysts. Returns FX ticks (bid/ask/spread), intraday OHLCV candles, crypto orderbook snapshots (depth 5-50), recent trades with VWAP, and sovereign bond yields. All sources are keyless public REST APIs (Binance, Coinbase, Kraken, OKX, open FX feeds, worldgovernmentbonds.com). Ultra-short cache: 10s for ticks/trades, 60s for orderbook. Use when an agent needs live market data as precise numeric inputs for trading logic, arbitrage detection, or portfolio valuation.","description_similarity":null,"before_hash":"5488e176689a066809b7af521a0ba517","after_hash":"5488e176689a066809b7af521a0ba517"},{"tool":"fx_rate","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Get the current or historical foreign-exchange rate for any currency pair — the exact exchange rate, FX rate or conversion rate an agent needs to convert a currency amount or feed a finance, trading, invoicing or pricing workflow. Covers EUR/USD, USD/JPY, GBP/EUR and every ISO-4217 currency pair. Returns the latest spot rate, or a historical rate by date. Use when a workflow needs a precise live or past currency exchange rate, or to convert money between two currencies. Source: European Central Bank reference rates via Frankfurter. Inputs: from/to ISO-4217 currency codes, optional date (YYYY-MM-DD).","description_similarity":null,"before_hash":"ee0470e5c74d42006115e38d9ac5c948","after_hash":"ee0470e5c74d42006115e38d9ac5c948"},{"tool":"working_capital","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Optimiseur du BFR — Gapup agent-payable C-suite expertise (CFO). Returns a structured, audited deliverable. Reference case: Agicap — BFR optimisation · DSO 52→38j · Cash libéré +€2.8M · 3 quick wins immédiats. Inputs are validated server-side — send the documented case fields.","description_similarity":null,"before_hash":"0c594a3e51c5959110f0eeb14c0a1729","after_hash":"0c594a3e51c5959110f0eeb14c0a1729"},{"tool":"win_loss_decoder","kind":"added","observed_at":"2026-09-30T14:38:08.055Z","fields":[],"description_before":null,"description_after":"Analyse Win/Loss deals — Gapup agent-payable C-suite expertise (CRO). Returns a structured, audited deliverable. Reference case: Gapup Hub — Win/Loss 32 deals Q1 2026 · Win rate 41% → 68% potentiel · Playbook 8 actions CRO. Inputs are validated server-side — send the documented case fields.","description_similarity":null,"before_hash":"754d78731bc582dc2a3b1d0b8fbaeea3","after_hash":"754d78731bc582dc2a3b1d0b8fbaeea3"}],"next_before":"224229","next_actions":[{"action":"list_changes","description":"Older changes","href":"/v1/servers/io.github.getgapup%2Fmcp-knowledge/changes?before=224229&limit=50"}]}