Context StudiosMCP Server
API key

Context Studios · Berlin · Model Context Protocol

359 tools. One server.

The MCP server behind contextstudios.ai. Agents read company facts, guides, glossary and comparisons without a key, and with a key they run the whole content, CRM and bookkeeping stack.

Kaiser-Friedrich-Str. 6 · 10585 BerlinMCP 2026-07-28 · 2025-11-25 · Streamable HTTP

Overview

Two endpoints. One catalog.

Tools in total
359
Public tools, no key
38
Public resources · prompts
3 · 3
Categories
21

Endpoints

Public or keyed. Same protocol.

Both endpoints speak JSON-RPC 2.0 over Streamable HTTP and hold no session state, so any request can land on any instance.

Public · no authentication(01)

POST https://mcp.contextstudios.ai/api/public/mcp

  • 38 tools, 3 resources, 3 prompts
  • Company facts, services, pricing model, blog, guides, glossary, comparisons, location and CMS pages
  • request_consultation is the only write: rate-limited, with a honeypot
  • search + fetch in the OpenAI deep-research shape for ChatGPT connectors
  • 60 requests per minute per IP
Authenticated · X-API-Key or OAuth 2.1(02)

POST https://mcp.contextstudios.ai/api

  • All 359 tools, with readOnlyHint, destructiveHint and openWorldHint annotations
  • X-API-Key header, or Authorization: Bearer from OAuth 2.1 (PKCE S256)
  • Dynamic client registration at POST /api/register
  • Destructive tools also require confirm: true
  • 300 requests per minute (OAuth), 600 (trusted API key)

Public surface

38 tools. No key.

Enough for any assistant to answer “What does Context Studios do?”, explain an AI term, compare two tools or estimate a project, and to hand over a consultation request.

(01)

Company & content8 tools

  • get_company_infoRetrieve comprehensive information about Context Studios including company overview, tech stack, differentiators, legal info, and contact details.
  • get_servicesList all 11 services offered by Context Studios.
  • get_pricing_modelHow Context Studios prices projects: fixed-price workshop entry from €1,500 (Light Discovery, Strategy Day, Prototyping Sprint), then scoped project pricing with a 48-hour proposal.
  • get_blog_postsSearch and retrieve published blog posts.
  • search_blog_knowledgeDeep full-text search through the entire blog knowledge base.
  • search_contentUniversal search across all public content.
  • ai_project_estimatorInteractive AI project effort estimator (S/M/L/XL, no prices).
  • blog_explorerInteractive blog browsing interface.
(02)

Guides3 tools

  • list_guidesList public guides.
  • get_guideGet a specific guide.
  • search_guidesSearch public guides.
(03)

Location pages3 tools

  • list_location_pagesList public location and bundesland pages.
  • get_location_pageGet specific location page.
  • search_location_pagesSearch public location and bundesland pages.
(04)

AI comparisons3 tools

  • get_comparisonRetrieve a detailed comparison between two AI tools.
  • list_comparisonsList all available comparisons.
  • search_comparisonsSearch comparisons by keyword.
(05)

AI glossary6 tools

  • get_glossary_termGet a detailed explanation of an AI term.
  • search_glossarySearch the AI glossary by keyword or category.
  • list_glossary_categoriesList all 10 glossary categories.
  • get_featured_glossary_termsGet a curated list of featured AI terms.
  • get_related_glossary_termsGet terms that are related to a specific glossary term.
  • get_glossary_by_tech_stackFilter glossary terms by technology provider or stack.
(06)

CMS pages2 tools

  • get_pageGet a specific CMS page by slug and locale.
  • search_pagesFull-text search across all published CMS pages.
(07)

Site content10 tools

  • get_faqGet frequently asked questions (site, or a specific service/industry).
  • list_industriesList the industries Context Studios serves.
  • get_industryGet an industry's full profile: TL;DR, entity definition, pain points, FAQ.
  • list_rolesList the buyer/user roles Context Studios addresses.
  • get_roleGet a role's full profile.
  • list_use_casesList the software use cases Context Studios builds.
  • get_use_caseGet a use case's full profile.
  • list_company_typesList the company-size segments Context Studios serves.
  • get_company_typeGet a company-size segment's full profile.
  • get_workshopsGet the three fixed-price entry workshops, booking link, and pricing statement.
(08)

Lead capture1 tool

  • request_consultationRequest a free consultation with Context Studios.
(09)

Deep research2 tools

  • searchSearch all public Context Studios content (blog, guides, glossary, AI comparisons).
  • fetchFetch a document by ID, or by a full contextstudios.ai URL/path — resolved to its Markdown twin.

Resources

(01)
  • company://profileStable company facts: services, stack, contact.
  • pricing://overviewHow pricing works: workshop entry from €1,500, project prices after scoping. And service packages.
  • mcp://public-toolsThe public catalog, derived from the live tool list.

Prompts

(02)
  • estimate_ai_projectGuides an agent from a project description to an estimate.
  • ask_context_studiosAnswers questions about the company with the right public tools.
  • compare_ai_toolsRoutes an “A vs B” question to the comparison tools.

Tool catalog

359 tools. 21 categories.

Every tool on the authenticated endpoint, grouped by job. Don’t know the name? Call search_tools with a keyword first.

Blog & translations27

Posts in DE, EN, FR, IT: write, translate, schedule, publish, verify live.

  • create_blog_postCreate a new blog post with full metadata, SEO tags, and multilingual support.Write
  • update_blog_postUpdate any aspect of an existing blog post including content, metadata, or SEO properties.Write
  • publish_blog_postPublish a draft blog post, making it live on the website.Destructive
  • translate_blog_postAI-powered translation of a blog post using Claude.Write
  • save_blog_translationsWrites EMBEDDED full-text translations into the source post document (legacy model) — it does NOT link standalone posts.Write
  • schedule_blog_postSchedule a draft to go live at a future time.Write
  • write_blog_contentAI-generate full blog post content from an outline.Write
  • list_blog_postsList blog posts with optional filters.Read
  • get_blog_postGet a single blog post by its slug.Read
  • unpublish_blog_postTake a published blog post offline.Destructive
  • delete_blog_postPermanently delete a blog post.Destructive
  • restore_blog_postSet a post back to draft status.Write
  • duplicate_blog_postCreate a copy of an existing blog post.Write
  • get_draft_postsList all draft posts.Read
  • get_draft_preview_urlGenerate a preview URL for a draft.Read
  • validate_blog_postChecks the two hero image fields only (featuredImageUrl, featuredImageMobileUrl) — NOT slug, meta description or content.Read
  • bulk_update_postsUpdate multiple posts at once.Write
  • check_translation_statusAudit translation completeness across the blog.Read
  • delete_blog_translationsRemove embedded translations from a post.Destructive
  • sync_translationsCross-link standalone posts that share the same slugs but different locales.Write
  • compare_translationsCompare a post with its translations for length parity and content completeness.Read
  • verify_blog_post_liveProves an article is actually live, not just written.Read
  • blog_slug_redirectsList or remove 301 rules for renamed blog slugs (blogSlugRedirects).Write
  • get_blog_analyticsGet analytics and statistics for blog posts including word counts, publishing frequency, and content metrics.Read
  • compare_postsCompare two blog posts side-by-side for metrics, SEO, and content analysis.Read
  • get_publishing_historyView recent publishing activity across blog and social media.Read
  • diagnose_publish_failureDebug why a post might have failed to publish.Read
Content pipeline9

Campaigns and end-to-end research → write → SEO → social pipelines, plus content stats.

  • create_campaignCreate a new content marketing campaign in Convex.Write
  • list_campaignsList all content campaigns from Convex.Read
  • get_campaign_statusGet detailed status and analytics for a campaign.Read
  • update_campaignUpdate campaign metadata or topic.Write
  • run_pipelineGet execution status of a content pipeline campaign.Write
  • run_researchRun the research step of the content pipeline.Write
  • run_pipeline_fullExecute complete content pipeline: research → outline → blog draft.Write
  • get_post_analyticsGet detailed traffic and engagement metrics for a specific post.Read
  • get_content_statsGlobal content metrics: total posts, published vs draft, locale distribution, recent activity.Read
Research8

Web research (Tavily), news, trends, knowledge base and content-gap analysis.

  • search_webPerform AI-powered web search using Tavily with intelligent answer summarization and source extraction.Read
  • research_topicComprehensive topic research combining multiple sources with configurable depth.Read
  • get_latest_newsGet the latest news articles on any topic with recency-focused results.Read
  • get_latest_infoDeep research combining Tavily web search with Gemini AI analysis for comprehensive, current information.Read
  • search_knowledge_baseSearch Context Studios' internal blog content and knowledge base.Read
  • suggest_topicsAI-generated blog topic suggestions based on trends, gaps, and audience interests.Read
  • get_trending_topicsGet currently trending topics in AI and technology.Read
  • analyze_content_gapsIdentify content gaps in the existing blog by analyzing coverage vs.Read
SEO & indexing13

Keywords, meta tags, internal links, SEO scores and Google Search Console submission.

  • generate_keywordsGenerate primary, secondary, and long-tail keywords for any topic.Write
  • generate_meta_tagsGenerate optimized SEO meta title and description.Write
  • generate_outlineGenerate SEO-optimized blog post outlines.Write
  • suggest_internal_linksAI suggests relevant internal linking opportunities.Read
  • inject_internal_linksAutomatically insert internal links into Markdown content.Write
  • analyze_seo_scoreComprehensive SEO analysis with score and recommendations using AI.Read
  • check_seo_healthBulk SEO analysis across all published blog posts.Read
  • submit_to_gscSubmit a URL to Google Search Console Indexing API.Write
  • check_gsc_index_statusCheck current indexing status in GSC.Read
  • bulk_submit_to_gscBatch submit multiple URLs to Google Indexing API.Write
  • batch_index_location_pagesBatch submit all city location pages.Write
  • batch_index_bundesland_pagesBatch submit all Bundesland pages.Write
  • optimize_seoRun SEO optimization on a blog post.Write
AEO telemetry4

Answer Engine Optimization: brand visibility across ChatGPT, Gemini and Perplexity.

  • audit_aeo_visibilityAudit brand visibility across AI assistants (ChatGPT, Gemini, Perplexity).Read
  • track_aeo_historyView visibility trends over time.Read
  • get_aeo_reportComprehensive report: mention rates per provider, best/worst queries, top competitors, actionable recommendations.Read
  • get_aeo_telemetryReport which AI agents (OpenAI, Anthropic, Perplexity, …) call the public MCP API: calls per agent, top tools, top agent+tool pairs, and top search queries.Read
Social publishing9

Posts for X, LinkedIn, Threads, Bluesky, Mastodon, Facebook and Instagram.

  • create_social_postAI-generate platform-optimized social media content from source material.Write
  • publish_to_socialPublish to multiple social platforms simultaneously via Typefully integration.Destructive
  • schedule_socialSchedule social media posts for future publishing at optimal times.Write
  • publish_to_facebookPublish directly to Facebook Page via Meta Graph API.Destructive
  • publish_to_instagramPublish to Instagram business account via Meta API.Destructive
  • get_social_statusGet the status of published or scheduled social media posts via Typefully.Read
  • list_social_setsList all Typefully social sets (accounts) available for publishing.Read
  • adapt_socialGenerate social media posts from a blog post for multiple platforms.Write
  • verify_blog_published_on_socialConfirm a blog slug actually appeared on Facebook and Instagram by counting needle matches in recent page/media posts.Read
Facebook & Instagram12

Meta Graph API: posts, comments, insights and moderation.

  • get_facebook_postsList recent posts from your Facebook Page with IDs, content, images, and metadata.Read
  • get_facebook_postGet details of a single Facebook post by its ID including message, image, and creation time.Read
  • delete_facebook_postPermanently delete a Facebook Page post.Destructive
  • update_facebook_postUpdate the message/content of an existing Facebook post.Write
  • get_facebook_post_insightsGet analytics for a specific Facebook post including impressions, reach, engagement, clicks, and reactions.Read
  • get_facebook_page_insightsGet page-level analytics including impressions, reach, engagement, and follower count over a time period.Read
  • get_instagram_mediaList recent posts from your Instagram Business account with IDs, captions, media types, and engagement metrics.Read
  • get_instagram_insightsGet analytics for a specific Instagram post including impressions, reach, engagement, and saves.Read
  • get_instagram_commentsList comments on an Instagram post with usernames, text, and timestamps.Read
  • delete_instagram_commentDelete a comment from an Instagram post.Destructive
  • reply_to_instagram_commentReply to a comment on an Instagram post.Write
  • toggle_instagram_commentsEnable or disable comments on an Instagram post.Write
Reddit6

Comment logging, rule checks, draft analysis and engagement patterns.

  • reddit_log_commentLog a posted Reddit comment with automatic style analysis.Write
  • reddit_update_scoresBatch update karma scores and reply counts for tracked Reddit comments.Write
  • reddit_get_rulesGet active engagement rules sorted by confidence.Read
  • reddit_analyze_draftPre-post quality check for Reddit comments.Write
  • reddit_daily_snapshotStore daily account-level metrics for Reddit.Write
  • reddit_pattern_analysisAnalyze last 14 days of comments to discover new patterns.Write
Images & vision11

Hero, blog and social images, screenshots and image analysis for text-only agents.

  • generate_hero_imageGenerate content-focused hero images using Gemini.Write
  • generate_blog_imageGenerate blog hero images at 2K.Write
  • generate_social_imageGenerate platform-optimized images with correct dimensions for each social network.Write
  • upload_image_to_storageUpload base64 image data to Convex storage.Write
  • list_image_stylesList all available image generation styles.Read
  • get_image_historyList blog posts with their featured images.Read
  • generate_imageGenerate any image from a text prompt.Write
  • take_screenshotCapture a high-quality retina screenshot of any website.Write
  • generate_hero_from_templateGenerate a branded hero image in the infographic design language (visual-design-systems.md): crafted SVG background motif (flow/editorial/spotlight/split/bento, …Write
  • generate_image_openrouterGenerate images via OpenRouter (default model: qwen/qwen-image-3-pro).Write
  • analyze_imageAnswer a question about an image (URL) using a vision model via OpenRouter.Read
Visual Studio2

Carousels, infographics, GIF animations, PDFs and mockups, plus visual analysis.

  • visual_studioVisual analysis AND creation for agents WITHOUT vision.Write
  • blog_to_carouselBridges a blogpost (the most expensive asset) directly to a validated LinkedIn carousel draft — no more hand-retyping slides from a post into create_carousel.Write
Video & audio21

Scripts, scenes, speech, lip sync, avatars, captions and post-production.

  • blog_to_shortFull blog-to-video-short pipeline.Write
  • generate_video_scriptGenerate a viral-formula video script from a topic or blog content.Write
  • generate_scene_promptsBreak a video script into scene-by-scene visual prompts for Veo 3.1.Write
  • generate_video_sceneGenerate a video scene using Google Veo 3.1.Write
  • check_video_generationPoll the status of a Veo 3.1 video generation.Write
  • generate_speechGenerate natural speech audio using ElevenLabs TTS (with-timestamps endpoint).Write
  • lipsync_videoSync lip movements in a video using Sync Labs.Write
  • transcribe_audioTranscribe audio files using OpenAI Whisper.Write
  • summarize_contentAI-powered content summarization concisely.Write
  • repurpose_contentTransform content for different platforms (e.g.Write
  • list_voicesList ElevenLabs voices.Write
  • check_lipsyncCheck status of a lip sync job.Write
  • check_video_statusCheck HeyGen video generation status.Read
  • check_video_creditsCheck remaining HeyGen video generation credits.Read
  • check_tts_usageCheck ElevenLabs subscription info and remaining character quota.Read
  • create_avatar_videoCreate a talking-head avatar video using HeyGen.Write
  • list_avatarsList available HeyGen avatars (talking head presenters) with IDs, names, and preview images.Read
  • get_lipsync_costEstimate the cost of a lip sync job before running it.Read
  • extend_video_sceneExtend an existing Veo 3.1 video by ~7 seconds.Write
  • generate_captions_srtGenerate estimated SRT captions from script text.Read
  • post_produce_videoGenerate a complete ffmpeg post-production script for video shorts.Read
Comparisons & glossary22

A-vs-B comparisons and glossary terms: CRUD, verified comparisons, AI generation.

  • create_comparisonCreate a new A vs B technology comparison.Write
  • update_comparisonUpdate an existing comparison.Write
  • delete_comparisonPermanently remove a comparison by ID or slug.Destructive
  • list_comparisonsList all available comparisons.Read
  • get_comparisonRetrieve a detailed comparison.Read
  • create_glossary_termCreate a new AI glossary term.Write
  • update_glossary_termUpdate an existing glossary term.Write
  • delete_glossary_termPermanently remove a glossary term by ID or slug.Destructive
  • list_glossary_termsList glossary terms.Read
  • get_glossary_termGet a detailed explanation of an AI term.Read
  • search_glossarySearch the AI glossary.Read
  • unpublish_comparisonTake a comparison offline (status back to draft, content is kept) and revalidate its pages.Write
  • unpublish_glossary_termTake a glossary term offline (status back to draft, content is kept) and revalidate its pages.Write
  • get_verified_comparisonRetrieve a verified evidence-backed comparison by slug and locale.Read
  • create_verified_comparisonWrite a new evidence-backed comparison to Convex (status=draft).Write
  • publish_verified_comparisonPublish a draft verified comparison.Write
  • rollback_verified_comparisonArchive a verified comparison, rolling back a publish or draft.Destructive
  • generate_comparisons_from_contentAI analyzes blog post or article content and generates structured A vs B comparisons.Write
  • suggest_comparisons_from_blogSuggest comparison topics based on existing blog content.Write
  • generate_glossary_from_contentAI extracts key AI/tech terms with definitions from content for glossary expansion.Write
  • suggest_glossary_from_blogSuggest new glossary terms from blog content.Write
  • enrich_glossary_termAI enriches an existing glossary term with related terms, tech stack associations, and use cases.Write
Guides5

Write access to the technical guides on contextstudios.ai/guides.

  • create_guideCreate or replace a guide (en/de).Write
  • update_guideReplace an existing guide's content (full Guide object in data).Write
  • publish_guidePublish a guide (status → published) and revalidate its pages.Write
  • unpublish_guideTake a guide offline (status → draft; content is kept) and revalidate its pages.Destructive
  • delete_guidePermanently delete one guide locale row and revalidate its pages.Destructive
Landing pages9

SEO landing pages: create, validate, publish, unpublish.

  • get_landing_pageGet a single landing page by slug and locale.Read
  • list_landing_pagesList all published landing pages.Read
  • create_landing_pageCreate a new landing page with meta data, sections, and optional structured data.Write
  • update_landing_pageUpdate an existing landing page's meta data, sections, or structured data.Write
  • publish_landing_pageSet a landing page's status to published, making it live on the site.Destructive
  • unpublish_landing_pageSet a published landing page back to draft.Destructive
  • delete_landing_pageArchive a landing page (soft delete).Destructive
  • validate_landing_pageValidate SEO/GEO requirements for a landing page.Read
  • generate_landing_page_contentGenerate a content template/structure for a landing page keyword.Write
CMS pages19

Marketing pages and their sections: edit, translate, diff, validate, sitemap.

  • list_cms_pagesList CMS pages with filters.Read
  • get_cms_pageGet a full CMS page by slug and locale.Read
  • create_cms_pageCreate a new CMS page.Write
  • update_cms_pageUpdate a CMS page's data.Write
  • delete_cms_pageArchive a CMS page.Destructive
  • duplicate_cms_pageClone a CMS page to another locale.Write
  • update_cms_sectionUpdate a single section within a CMS page.Write
  • add_cms_sectionAdd a new section to a CMS page.Write
  • remove_cms_sectionRemove a section from a CMS page.Write
  • reorder_cms_sectionsReorder sections within a CMS page.Write
  • translate_cms_pageAuto-translate a CMS page.Write
  • search_cms_pagesFull-text search across CMS pages.Read
  • validate_cms_pageCheck a CMS page for SEO completeness.Read
  • diff_cms_pageCompare a CMS page across two locales.Read
  • preview_cms_pageGenerate a preview URL.Read
  • get_cms_statsOverview statistics: page counts.Read
  • generate_cms_sitemapGenerate sitemap entries for all published CMS pages.Read
  • bulk_create_cms_pagesCreate multiple CMS pages at once.Write
  • bulk_update_cms_metaMass-update SEO metadata.Write
Outreach12

Outreach leads, campaigns and pipeline runs.

  • create_leadCreate a new lead in the outreach CRM.Write
  • update_leadUpdate lead information.Write
  • list_leadsList all leads with filters.Read
  • search_leadsSearch leads by keyword.Read
  • update_lead_statusUpdate lead pipeline status.Write
  • add_lead_noteAdd a note to a lead.Write
  • get_outreach_statsGet outreach performance and pipeline stats.Read
  • delete_leadDelete a lead.Destructive
  • create_campaign_outreachCreate an outreach campaign.Write
  • list_campaigns_outreachList all outreach campaigns.Read
  • get_campaign_status_outreachGet outreach campaign status and metrics.Read
  • run_pipeline_outreachExecute an outreach campaign pipeline action.Write
CRM, offers & invoices57

Leads, clients, offers, invoices, guided wizards and the deal pipeline.

  • crm_health_checkProbe the Convex backend to verify CRM/NDA/wizard tools are reachable.Read
  • crm_list_leadsList CRM leads with filters.Read
  • crm_get_leadGet lead details and activities history.Read
  • crm_get_lead_statsPipeline overview for leads.Read
  • crm_list_clientsList clients or vendors with status or search term.Read
  • crm_get_clientGet details of a specific client.Read
  • crm_list_offersFilter offers by client, lead, or status.Read
  • crm_get_offerGet details and version history of an offer.Read
  • crm_get_offer_statsOffer pipeline statistics.Read
  • crm_list_offer_templatesList title and item templates for offers.Read
  • crm_list_offer_clausesList reusable bilingual prose clauses for Workshop/Project offer sections.Read
  • crm_export_offer_pdfGenerate professional offer PDF.Read
  • crm_export_invoice_pdfGenerate professional invoice PDF.Read
  • crm_list_invoicesList invoices with filters.Read
  • crm_get_invoiceGet details of a specific invoice.Read
  • crm_get_offer_wizard_statusReturn the current WizardEnvelope for an open offer session: sessionId, type, status, currentStep, and completeness.Read
  • crm_preview_offer_wizardReturn a WizardEnvelope with a propsSummary snapshot of the offer session (counterparty, line items, totals, sections) without generating a PDF.Read
  • crm_list_offer_wizard_sessionsList open (or submitted/discarded) offer wizard sessions.Read
  • crm_get_invoice_wizard_statusReturn the current WizardEnvelope for an open invoice session: sessionId, type, status, currentStep, and completeness.Read
  • crm_preview_invoice_wizardReturn a WizardEnvelope with a propsSummary snapshot of the invoice session (client, line items, totals, tax) without generating a PDF.Read
  • crm_list_invoice_wizard_sessionsList open (or submitted/discarded) invoice wizard sessions.Read
  • lws_get_lead_wizard_statusGet the current state and completeness of a lead wizard session by sessionId.Read
  • lws_preview_lead_wizardGet a structured summary of all collected lead data before finalizing.Read
  • lws_list_lead_wizard_sessionsList lead wizard sessions filtered by status.Read
  • pipeline_get_statusGet a summary of the current deal pipeline status for a lead: what phase they're in (new → qualified → proposal → negotiation → closed_won/lost), what's outstanding …Read
  • pipeline_list_activeList all active leads with their pipeline phase (offer pending, offer sent, invoice pending, waiting for payment).Read
  • pipeline_suggest_next_actionAnalyze a lead's full timeline (notes, activities, offers, invoices) and suggest the single most important next action to advance the deal.Read
  • pipeline_get_statsGet aggregated pipeline statistics: total leads by status, conversion rates, estimated pipeline value.Read
  • crm_create_leadCreate a new lead in the CRM.Write
  • crm_update_leadUpdate lead fields. Pass all fields to update as top-level properties (not nested in 'updates').Write
  • crm_add_lead_noteAdd a note, email, call, or meeting to a lead.Write
  • crm_convert_lead_to_clientIRREVERSIBLE: Convert lead to client and migrate history.Destructive
  • crm_create_clientCreate a new customer or vendor.Write
  • crm_update_clientUpdate client/vendor fields.Write
  • crm_create_offerCreate a new offer for a client/lead.Write
  • crm_update_offerEdit a draft offer.Write
  • crm_send_offerSend an offer to a recipient by email.Write
  • crm_create_offer_versionCreate a new version of an offer.Write
  • crm_add_offer_sectionAdd a custom Markdown prose section to a draft offer (e.g.Write
  • crm_update_offer_sectionUpdate an existing section on a draft offer.Write
  • crm_remove_offer_sectionRemove a section from a draft offer.Destructive
  • crm_attach_offer_clauseAttach a clause from the library to a draft offer as a new section.Write
  • crm_apply_offer_title_templateApply a title-template's clause-based section set to a draft offer.Write
  • crm_set_offer_section_variablesSet/merge {{variable}} substitution values on an offer.Write
  • crm_delete_smoke_leadDelete a smoke-test lead (source must start with '_smoke_').Destructive
  • crm_start_offer_wizardOpen a new guided offer-creation session.Write
  • crm_answer_offer_wizardUpdate an open offer wizard session with field values.Write
  • crm_create_offer_from_wizardFinalize a completed offer wizard session and create a real CRM offer record.Write
  • crm_discard_offer_wizardAbandon an open offer wizard session.Destructive
  • crm_start_invoice_wizardOpen a new guided invoice-creation session.Write
  • crm_answer_invoice_wizardUpdate an open invoice wizard session with field values.Write
  • crm_create_invoice_from_wizardFinalize a completed invoice wizard session and create a real CRM invoice record with an auto-assigned invoice number (RE-YYYY-NNNN).Write
  • crm_discard_invoice_wizardAbandon an open invoice wizard session.Destructive
  • lws_start_lead_wizardOpen a new guided lead-qualification session.Write
  • lws_answer_lead_questionUpdate an open lead wizard session with field values.Write
  • lws_discard_lead_wizardDiscard an open lead wizard session without creating a lead.Destructive
  • lws_create_lead_from_wizardFinalize a complete lead wizard session and create the CRM lead.Write
Accounting & disclosure25

SKR03 bookkeeping, ELSTER and DATEV exports, payment wizard, annual disclosure.

  • get_accounting_balancesGet current SKR03 account balances (1200, 1410, 1576/1577, 1787, 0880).Read
  • list_open_expensesList all expenses without journal entries.Read
  • generate_elster_xmlGenerate ELSTER-compliant UStVA XML.Read
  • get_financial_runwayCalculate remaining months based on Qonto balance and burn rate.Read
  • export_datevExport bookings as DATEV-compliant CSV.Read
  • acc_get_payment_wizard_statusGet the current state and completeness of a payment wizard session by sessionId.Read
  • acc_preview_payment_wizardGet a structured summary of the payment to be recorded: invoice number, total, payment amount, date, and method.Read
  • acc_list_payment_wizard_sessionsList payment wizard sessions filtered by status.Read
  • book_expenseCreate an expense and corresponding journal entry.Write
  • match_incoming_paymentLink a Qonto transaction with an open invoice.Write
  • classify_qonto_transactionsAI classification of pending Qonto transactions.Write
  • book_manual_journalCreate a free-form SKR03 journal entry (multi-line debit/credit).Write
  • delete_expenseDelete an expense and its linked journal entry.Destructive
  • correct_expenseCorrect an existing expense using the German Storno+Neubuchung pattern: cancels the old expense, creates a reversal journal entry, then books a new corrected expense …Destructive
  • delete_journal_entryDelete a standalone journal entry (not linked to an expense).Destructive
  • acc_start_payment_wizardOpen a guided payment-recording session.Write
  • acc_answer_payment_questionUpdate an open payment wizard session with field values.Write
  • acc_discard_payment_wizardDiscard an open payment wizard session without recording a payment.Destructive
  • acc_record_payment_from_wizardFinalize a complete payment wizard session: records the payment in the payments table, updates invoice.status (paid if amount >= total), and creates a GoBD-compliant …Write
  • preview_disclosureBilanz-Vorschau für die Hinterlegung beim Unternehmensregister.Read
  • list_disclosure_filingsListet alle Unternehmensregister-Hinterlegungen mit Audit-Trail (draft → generated → validated → submitted → confirmed).Read
  • export_disclosure_xmlLiefert den vollständigen XML-Inhalt einer Hinterlegung zurück.Read
  • generate_disclosure_filingErzeugt eine neue Unternehmensregister-Hinterlegung für das angegebene GJ.Write
  • mark_disclosure_submittedMarkiert eine Hinterlegung als beim Unternehmensregister eingereicht.Write
  • mark_disclosure_confirmedMarkiert eine submitted-Hinterlegung als vom Unternehmensregister bestätigt (nach Erhalt der Eingangsbestätigung).Write
NDA lifecycle36

Templates, clauses, drafts, signatures, PDF rendering and revisions.

  • nda_start_wizardStart an NDA generation session.Write
  • nda_answer_questionUpdate an open NDA wizard session with answers and return the next missing fields.Write
  • nda_previewReturn a structured preview for an open NDA session: completeness state, variables, clause titles, and PDF URL if a draft has already been rendered.Read
  • nda_get_statusGet the current state for an NDA wizard session.Read
  • nda_listList existing NDAs from Convex with optional status/search filtering.Read
  • nda_getGet one existing NDA with signers, edit requests, recent audit events, and optionally signed PDF URLs.Read
  • nda_app_list_templatesApp-only helper that returns built-in NDA templates.Read
  • nda_app_update_sessionApp-only helper for synchronizing NDA wizard state.Write
  • nda_app_search_partiesApp-only helper that searches the CRM for clients and leads matching the query string.Read
  • nda_app_render_previewApp-only helper. Returns a structured preview of an NDA (clause titles + a signed temporary URL the iframe can fetch directly).Read
  • nda_render_unsigned_pdfRender or re-render the unsigned NDA PDF and return a signed temporary URL.Write
  • nda_app_read_pdf_chunkApp-only helper. Returns a signed URL to one of the stored PDF kinds (unsigned, user-signed, fully-executed).Read
  • nda_list_clausesList built-in and custom NDA clauses.Read
  • nda_list_templatesList NDA templates. By default returns all templates (built-in + custom).Read
  • nda_exportExport a complete NDA with metadata, signers, audit events, edit requests, resolved clauses, and optional signed PDF URLs.Read
  • nda_get_wizard_statusAlias for nda_get_status.Read
  • nda_preview_wizardAlias for nda_preview.Read
  • nda_list_wizard_sessionsList open (or completed/abandoned) NDA wizard sessions.Read
  • nda_create_draftCreate a Convex NDA draft from a completed NDA wizard session.Write
  • nda_send_for_signatureSend an NDA signing link.Write
  • nda_update_party_detailsUpdate our signer and counterparty details before sending.Write
  • nda_change_statusChange an NDA status using the same lifecycle rules as the admin app.Destructive
  • nda_deletePermanently delete an NDA, related signers/edit requests/access challenges, and stored PDFs/signatures.Destructive
  • nda_render_final_pdfRender the fully executed NDA PDF and return a signed temporary URL.Write
  • nda_retry_final_renderRetry a failed final PDF render by scheduling the same internal render job as the admin app.Write
  • nda_app_sendApp-only helper for confirmed NDA sends.Write
  • nda_prepare_user_signatureStores the admin's signature image (base64 PNG) on the 'us' signer of an NDA, then re-renders the user-signed PDF.Write
  • nda_app_save_signatureApp-only variant of nda_prepare_user_signature.Write
  • nda_create_revision_from_edit_requestClones an NDA into a new draft revision in response to a counterparty edit request.Write
  • nda_create_clauseCreate a custom NDA clause in Convex.Write
  • nda_update_clauseUpdate a custom NDA clause by Convex id or custom key.Write
  • nda_upsert_clauseCreate or replace a custom NDA clause by key.Write
  • nda_create_templateCreate a custom NDA template with ordered clause keys and default variables.Write
  • nda_update_templateUpdate a custom NDA template.Write
  • nda_create_from_wizardAlias for nda_create_draft.Write
  • nda_discard_wizardAbandon an open NDA wizard session.Destructive
Cortex & Nightshift14

Long-term agent memory (Cortex) and multi-agent coding orchestration (Nightshift).

  • cortex_rememberStore a memory with auto-embedding, auto-association, contradiction detection, and dual-write to markdown fallback.Write
  • cortex_recallSemantic search with spreading activation.Read
  • cortex_list_todayList all memories created since a given timestamp (milliseconds).Read
  • cortex_what_do_i_knowKnowledge audit on a topic.Read
  • cortex_why_did_weDecision tracing via multi-hop association traversal.Read
  • cortex_forgetSoft-delete/archive a memory.Destructive
  • cortex_statsSystem health dashboard: total memories, breakdown by store/category.Read
  • cortex_checkpointSave current working context for session recovery.Write
  • cortex_wakeResume from a checkpoint with hydrated memories.Read
  • nightshift_startStart a nightshift session for a specific agent and spec file.Write
  • nightshift_sendSend a message or instruction to a running nightshift agent.Write
  • nightshift_readRead pending messages from nightshift agents (drains the queue).Read
  • nightshift_statusGet real-time status of all nightshift agents and Hub health.Read
  • nightshift_stopStop a running nightshift agent.Write
Public, apps & discovery42

The 38 public tools, the MCP Apps and tool search.

  • get_company_infoRetrieve comprehensive information about Context Studios including company overview, tech stack, differentiators, legal info, and contact details.Read
  • get_servicesList all 11 services offered by Context Studios.Read
  • get_pricing_modelHow Context Studios prices projects: fixed-price workshop entry from €1,500, then scoped project pricing with a 48-hour proposal.Read
  • get_blog_postsSearch and retrieve published blog posts.Read
  • search_blog_knowledgeDeep full-text search through the entire blog knowledge base.Read
  • search_contentUniversal search across all public content.Read
  • ai_project_estimatorInteractive AI project effort estimator (S/M/L/XL, no prices).Read
  • blog_explorerInteractive blog browsing interface.Read
  • list_guidesList public guides.Read
  • get_guideGet a specific guide.Read
  • search_guidesSearch public guides.Read
  • list_location_pagesList public location and bundesland pages.Read
  • get_location_pageGet specific location page.Read
  • search_location_pagesSearch public location and bundesland pages.Read
  • get_comparisonRetrieve a detailed comparison between two AI tools.Read
  • list_comparisonsList all available comparisons.Read
  • search_comparisonsSearch comparisons by keyword.Read
  • get_glossary_termGet a detailed explanation of an AI term.Read
  • search_glossarySearch the AI glossary by keyword or category.Read
  • list_glossary_categoriesList all 10 glossary categories.Read
  • get_featured_glossary_termsGet a curated list of featured AI terms.Read
  • get_related_glossary_termsGet terms that are related to a specific glossary term.Read
  • get_glossary_by_tech_stackFilter glossary terms by technology provider or stack.Read
  • get_pageGet a specific CMS page by slug and locale.Read
  • search_pagesFull-text search across all published CMS pages.Read
  • get_faqGet frequently asked questions (site, or a specific service/industry).Read
  • list_industriesList the industries Context Studios serves.Read
  • get_industryGet an industry's full profile: TL;DR, entity definition, pain points, FAQ.Read
  • list_rolesList the buyer/user roles Context Studios addresses.Read
  • get_roleGet a role's full profile.Read
  • list_use_casesList the software use cases Context Studios builds.Read
  • get_use_caseGet a use case's full profile.Read
  • list_company_typesList the company-size segments Context Studios serves.Read
  • get_company_typeGet a company-size segment's full profile.Read
  • get_workshopsGet the three fixed-price entry workshops, booking link, and pricing statement.Read
  • request_consultationRequest a free consultation with Context Studios.Write
  • searchSearch all public Context Studios content (blog, guides, glossary, AI comparisons).Read
  • fetchFetch the full content of a document by ID returned from the search tool.Read
  • ai_project_estimatorEstimate the cost and timeline for an AI/software project.Write
  • blog_explorerInteractive blog explorer for Context Studios.Write
  • search_toolsFind the right tool by keyword instead of scanning the full catalog.Read
  • submit_leadLead capture called from the MCP App iframes (project estimator, blog explorer).Write

Counts are unique tool names. Six read tools are listed in two groups below: four comparison and glossary readers, and the two MCP App tools, which are also public. Write tools for accounting, CRM, NDAs and verified comparisons are switched on per deployment.

Guides · write access

Publish a guide. No deploy.

The guides on contextstudios.ai/guides live in the database. Five authenticated tools write them, and every published change revalidates the site and the edge cache.

  1. (01)create_guideCreate or replace a guide (en/de). New guides are drafts unless status='published'; publish later with publish_guide. data must be the full Guide object. Published writes trigger on-demand revalidation (no deploy).
  2. (02)update_guideReplace an existing guide's content (full Guide object in data). Keeps its current status (draft stays draft, live stays live) unless status is given. Revalidates the site when the guide is (or was) published.
  3. (03)publish_guidePublish a guide (status → published) and revalidate its pages. Fails with the list of blockers when the guide is incomplete (kebab slug, tldr, ≥3 faqs, a ranking/criteria/self-test/steps body).
  4. (04)unpublish_guideTake a guide offline (status → draft; content is kept) and revalidate its pages. A DE draft falls back to the EN guide on /de/guides/<slug>.
  5. (05)delete_guidePermanently delete one guide locale row and revalidate its pages. Requires confirm: true. Prefer unpublish_guide to keep the content.

Protocol status

2026-07-28 live. 2025-11-25 too.

What the server does today, checked against the spec and client probes on 2026-09-23. Built on the MCP TypeScript SDK v2: one endpoint, both revisions.

Current(01)

MCP 2026-07-28

Stateless per request: the _meta envelope, server/discover, resultType, cache hints and the Mcp-Method/Mcp-Name header checks. Clients that negotiate it get it automatically.

Supported(02)

MCP 2025-11-25

The revision Claude Code, Claude Desktop, Cursor and ChatGPT connectors use with initialize. Served on the same endpoints, unchanged: tools, resources, prompts, progress notifications and MCP Apps.

FeatureStatusNotes
Stateless Streamable HTTPLiveNo Mcp-Session-Id, no handshake required; a fresh server per request.
server/discoverLiveBoth endpoints; lists 2026-07-28 and 2025-11-25.
Mcp-Method / Mcp-Name headersLiveRequired on 2026-07-28 requests; a mismatch returns -32020.
ttlMs / cacheScopeLiveOn all list results and resources/read; 60 s, public on /api/public/mcp, private on /api.
resultType, error -32022LiveEvery 2026-07-28 result carries resultType; unsupported versions get -32022.
subscriptions/listenNot offeredCatalogs are static (listChanged: false); the method returns -32601.
Origin validationLog-onlyUnlisted origins are logged, not blocked; auth is key or bearer, never cookies.
OAuth 2.1, PKCE S256LiveRFC 8414 and RFC 9728 metadata, RFC 7591 registration, refresh-token rotation, audience check.
Client ID Metadata DocumentsOpenDynamic client registration works today.
Tool annotations, outputSchema, iconsLiveAll public tools carry title, annotations, outputSchema and light/dark category icons.
Resource templatesLiveglossary://, comparison://, guide://, industry://, role://, use-case://, company-type:// + {slug}.
completion/completeLiveSlug suggestions for every template, locales for prompt arguments.
Audit trailLiveEvery call is logged with a hashed credential fingerprint and redacted arguments.

Connect

One line in your client. Or plain curl.

Any MCP client with Streamable HTTP works: Claude Code, Claude Desktop, Cursor, Windsurf, VS Code, ChatGPT.

Claude Code · publicbash
claude mcp add --transport http context-studios \
  https://mcp.contextstudios.ai/api/public/mcp
Claude Code · with keybash
claude mcp add --transport http context-studios \
  https://mcp.contextstudios.ai/api \
  --header "X-API-Key: $MCP_API_KEY"
.mcp.json / claude_desktop_config.jsonjson
{
  "mcpServers": {
    "context-studios-public": {
      "command": "npx",
      "args": ["mcp-remote", "https://mcp.contextstudios.ai/api/public/mcp"]
    },
    "context-studios": {
      "command": "npx",
      "args": ["mcp-remote", "https://mcp.contextstudios.ai/api", "--header", "X-API-Key:${MCP_API_KEY}"]
    }
  }
}
Direct HTTP · no keybash
# List the public tools
curl -X POST https://mcp.contextstudios.ai/api/public/mcp \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -d '{"jsonrpc":"2.0","method":"tools/list","id":1}'

# Call one
curl -X POST https://mcp.contextstudios.ai/api/public/mcp \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -d '{"jsonrpc":"2.0","method":"tools/call","params":{"name":"get_comparison","arguments":{"slug":"claude-code-vs-github-copilot"}},"id":2}'

Headers and limits

Header / limitWhenValue
Content-TypeAlwaysapplication/json
AcceptAlwaysapplication/json, text/event-stream
X-API-Key/apiYour API key
Authorization/api via OAuthBearer <token>
MCP-Protocol-VersionOptional2026-07-28 with matching Mcp-Method (and Mcp-Name for calls); or 2025-11-25
Rate limit, public/api/public/mcp60 per minute per IP
Rate limit, keyed/api300 per minute (OAuth), 600 (trusted API key)

API access

Need the full catalog? Talk to us.

API keys go to clients and partners. We set up scopes, rate limits and the write tools you need.

  1. (01)
    Tell us the workflow

    Which tools, which data, which agent.

  2. (02)
    Get a key or an OAuth client

    Scoped to your use, revocable at any time.

  3. (03)
    Connect and ship

    One line in Claude Code or any MCP client.