Heista is an MCP server with 118 tools an agent can call, among them add_brand_asset, adformula_intelligence, call_creative_agent_preset. It answers at https://www.heista.co/api/mcp/mcp over streamable HTTP, and it listed its tools to an anonymous client, so no sign-in was needed for that. Our probes reached it 10 of 10 times since 4 September 2026; the last check, on 17 October 2026, succeeded.
Decode video ads, load brand intelligence, generate ad scripts.
Description as the publisher wrote it.
| Tool | What it does |
|---|---|
add_brand_asset | Upload an image to a brand by URL. The pipeline downloads it, runs the vision tagger (classifies type, detects product name, flags is_primary_product), stores it in the brand-assets bucket, and inserts a brand_assets row. Paid (vision tag c Inputs: brand_id, image_url |
adformula_intelligence | Browse proven ad formula blueprints — structural patterns clustered from 3-10+ winning ads that independently converged on the same beat architecture while Meta kept rewarding them with sustained spend. Takes optional filters: vertical, cre Inputs: vertical, creative_format, marketing_angle, algo_intent, hook_type, limit |
call_creative_agent_preset | Invoke a Creative Agent (character) preset. Every preset is a purpose-built character the workspace has authored or the Heista catalog has published — copy voice, art direction, strategy, creative direction, etc. ONE-SHOT: give the characte Inputs: agent_id, message, thread_id, brand_id, working_context, idempotency_key |
call_creative_worlds | Heista's creative direction engine — same engine the Creative Director specialist runs internally, exposed over MCP. ONE-SHOT: give a brief, get N finished creative outputs. For back-and-forth refinement, or output shapes the `medium` enum Inputs: brief, count, brand_id, medium, lens_hint, idempotency_key |
chat_with_creative_worlds | Multi-turn conversation with Heista's creative direction engine — a real chat where the agent decides each turn what to produce based on what you ask for. Use whenever the work needs more than one round, OR when you want an output shape not Inputs: message, session_id, brand_id |
check_balance | Check the calling user's Heista API credit balance, month-to-date usage broken down by operation, lifetime spend, and the current pricing for every paid tool. Takes no inputs. Returns balance in cents, lifetime spend in cents, month-to-date |
create_powersource_docs | Build a complete creative intelligence profile from internal brand documents — creative briefs, brand guidelines, product specs, customer research, competitive analysis. Takes any mix of file_ids (from a previous upload), document_urls (pub Inputs: brand_id, file_ids, document_urls, documents_inline, context_url, idempotency_key |
create_powersource_full | Build the highest-fidelity creative intelligence profile by combining a brand's public website URL with their internal documents. Takes a required website URL plus at least one document — file_ids from previous upload, public document_urls Inputs: url, brand_id, file_ids, document_urls, documents_inline, idempotency_key |
create_powersource_url | Build a complete creative intelligence profile of a brand from a single website URL. Takes a website URL (homepage, PDP, landing page) plus optional idempotency_key, force_refresh, and webhook_url. Returns a job_id immediately; poll with ge Inputs: url, brand_id, idempotency_key, force_refresh, webhook_url |
creative_delete_draft | Permanently delete a Creative Library draft that has NEVER been published. Refused for any article with publication history — use creative_unpublish_article to take a live page down instead, which keeps the draft. Use this only to clear tes Inputs: shelf, slug |
creative_get_authoring_contract | Read the canonical shelves, taxonomy, relationship Lego, safe content rules, and positional slots for every publishable Creative Library format. Call this before authoring. |
creative_get_draft | Read one Creative Library draft so it can be reviewed before explicit publication. Inputs: shelf, slug |
creative_get_shelf | Read one shelf hub as an object: its public URL, what it covers, what it deliberately does NOT cover (and which shelf owns each of those instead), and every article on it with status and live URL. Use it to check placement before drafting. Inputs: shelf |
creative_list_articles | List Creative Library drafts and publication state. Returns editorial metadata only, not customer data. Inputs: limit, offset |
creative_publish_article | Publish only the exact reviewed draft revision. If the draft changed after review, publication fails and the new revision must be reviewed. Inputs: shelf, slug, expected_revision |
creative_save_draft | Validate and save a complete Creative Library draft using the canonical taxonomy and format-specific content slots. An identical retry is a no-op. It cannot replace published content or publish a page. |
creative_unpublish_article | Take a live Creative Library article down. It stops being public and leaves the sitemap immediately. The draft is preserved and can be re-published after review. Use this to reverse a publication. Inputs: shelf, slug |
creative_update_slug | Change an article slug and 301 the old URL at the new one. Never leaves two live URLs. Refuses a new slug ending in a year, a slug already in use, and any article whose draft is ahead of what is published (renaming republishes). Use this in Inputs: shelf, slug, new_slug |
decode_ad | Decode a specific video ad URL into its full structural formula — beat-by-beat breakdown, hook classification, behavioral psychology stack, creative format, runtime performance signals (active days on Meta Ad Library when available), and pe Inputs: url, idempotency_key |
decoder_intelligence | Browse individual decoded ads from Heista's corpus of real winning Meta/TikTok creative. Takes optional filters: vertical, creative_format, marketing_angle, hook_type, algo_intent, brand (partial name match), and limit (1-10, default 5). Ea Inputs: vertical, creative_format, marketing_angle, hook_type, algo_intent, brand |
delete_brand_asset | Delete one brand asset by asset_id. Removes the brand_assets row and (when the asset was uploaded rather than scanned) the storage object. Destructive — confirm with the user before calling. Use list_brand_assets first to find the asset_id. Inputs: asset_id |
delete_saved_asset | Delete one saved asset by id. Destructive — confirm with the user before calling. OAuth callers can only delete saves they created themselves (Linear model — see /assets UI for org-admin override). API-key callers are treated as org-trusted Inputs: asset_id |
dispatch_desk_researcher | Dispatch to the DESK RESEARCHER — source-grounded synthesis on a topic landscape. Use for: "what is known about X / give me the landscape of Y / fact-check Z / synthesize the published evidence on W". Multi-source FACT/INFERENCE extraction Inputs: objective, output_format, tool_guidance, boundaries, priority |
dispatch_desk_researcher_async | Dispatch to the DESK RESEARCHER — source-grounded synthesis on a topic landscape. Use for: "what is known about X / give me the landscape of Y / fact-check Z / synthesize the published evidence on W". Multi-source FACT/INFERENCE extraction Inputs: objective, output_format, tool_guidance, boundaries, priority |
dispatch_head_of_research | Run a full research workflow via the Head of Research agent. The Head decomposes your brief into specialist sub-questions, dispatches the right combination of 6 specialists (desk, trend, market, quant, qual, social) in parallel via async di Inputs: brief, priority, max_wait_seconds |
dispatch_market_analyst | Dispatch to the MARKET ANALYST — entity-deep teardown of a named brand or vendor. Use for: "what is brand X / how does company Y work / decode competitor Z / teardown vendor W". Multi-axis extraction grounded in multi-class sourcing, plus d Inputs: objective, output_format, tool_guidance, boundaries, priority |
dispatch_market_analyst_async | Dispatch to the MARKET ANALYST — entity-deep teardown of a named brand or vendor. Use for: "what is brand X / how does company Y work / decode competitor Z / teardown vendor W". Multi-axis extraction grounded in multi-class sourcing, plus d Inputs: objective, output_format, tool_guidance, boundaries, priority |
dispatch_qualitative_researcher | Dispatch to the QUALITATIVE RESEARCHER — thematic synthesis from unstructured text (interviews, reviews, forum threads, customer language). Use for: "what are the 2-3 recurring themes in how D2C founders talk about X / what language is bein Inputs: objective, output_format, tool_guidance, boundaries, priority |
dispatch_qualitative_researcher_async | Dispatch to the QUALITATIVE RESEARCHER — thematic synthesis from unstructured text (interviews, reviews, forum threads, customer language). Use for: "what are the 2-3 recurring themes in how D2C founders talk about X / what language is bein Inputs: objective, output_format, tool_guidance, boundaries, priority |
dispatch_quantitative_researcher | Dispatch to the QUANTITATIVE RESEARCHER — numerical analysis with full methodology context. Use for: briefs that turn on numbers done rigorously — "what is the documented effect size of X / what does the data say about Y / quantify the impa Inputs: objective, output_format, tool_guidance, boundaries, priority |
dispatch_quantitative_researcher_async | Dispatch to the QUANTITATIVE RESEARCHER — numerical analysis with full methodology context. Use for: briefs that turn on numbers done rigorously — "what is the documented effect size of X / what does the data say about Y / quantify the impa Inputs: objective, output_format, tool_guidance, boundaries, priority |
dispatch_social_listening_researcher | Dispatch to the SOCIAL LISTENING RESEARCHER — multi-platform community-signal interpretation. Use for: "what are practitioners saying about X across platforms / what jargon is emerging in field Y / what is the cross-platform discourse aroun Inputs: objective, output_format, tool_guidance, boundaries, priority |
dispatch_social_listening_researcher_async | Dispatch to the SOCIAL LISTENING RESEARCHER — multi-platform community-signal interpretation. Use for: "what are practitioners saying about X across platforms / what jargon is emerging in field Y / what is the cross-platform discourse aroun Inputs: objective, output_format, tool_guidance, boundaries, priority |
dispatch_trend_researcher | Dispatch to the TREND RESEARCHER — recency-dominant trajectory investigation. Use for: "is X a real trend / what is happening with X right now / where is X headed / what is driving X". Distinguishes trend from spike, signal from noise, real Inputs: objective, output_format, tool_guidance, boundaries, priority |
dispatch_trend_researcher_async | Dispatch to the TREND RESEARCHER — recency-dominant trajectory investigation. Use for: "is X a real trend / what is happening with X right now / where is X headed / what is driving X". Distinguishes trend from spike, signal from noise, real Inputs: objective, output_format, tool_guidance, boundaries, priority |
favorite_saved_asset | Toggle the favorite flag on a saved asset. Pass is_favorite=true to favorite, false to unfavorite. favorited_at is set/cleared in lockstep so the Favorites tab sorts correctly. Not destructive. Inputs: asset_id, is_favorite |
fetch_url | Drill into a specific URL after search surfaces it. Returns the extracted text content plus metadata. Internal routing: PDFs hit Anthropic Files API for OCR + structured extraction; HTML pages are fetched + text-extracted via readability-st Inputs: url, max_chars |
fleet_analytics_overview | Aggregate marketing analytics for the last 7/28/90 days: pageviews, visitors, sessions, AI-search-referred sessions, the view→engaged→CTA→signup→trial funnel, and top pages. Aggregates only — never person-level data. FUNNEL EVENTS, so you k Inputs: days |
fleet_analytics_top_pages | Top marketing pages by views for the last 7/28/90 days, optionally filtered to a path prefix (e.g. "/decode", "/intelligence", "/brands"). Includes engagement signals where captured. Aggregates only, hard cap 50 rows. Read-only. Inputs: days, path_prefix, limit |
fleet_analytics_trend | Daily pageview series for the last 7/28/90 days, split by traffic source category (ai_search / organic / social / direct / referral). Use to measure launch weeks and content momentum. Aggregates only. Read-only. Inputs: metric, days |
fleet_crawler_hits, fleet_create_issue, fleet_get_brand_report, fleet_get_decoded_ad, fleet_get_issue, fleet_gsc_inspect_url, fleet_gsc_query, fleet_gsc_sitemaps, fleet_gsc_summary, fleet_gsc_top_pages, fleet_gsc_top_queries, fleet_intel_stats, fleet_list_brand_reports, fleet_list_intelligence_articles, fleet_list_issues, fleet_product_funnel_counts, fleet_product_funnel_summary, fleet_product_signups_recent, fleet_product_user_summary, fleet_search_decoded_ads, fleet_seo_recovery, fleet_site_audit_summary, fleet_site_links, fleet_site_pages, generate_adscript, get_ad_formula_preset, get_brand, get_cd_card_bookmark_preset, get_creative_agent_preset, get_creative_agent_skill_preset, get_creative_director_playbook_preset, get_decode, get_decoded_ad_preset, get_dispatch_result, get_fleet_cost, get_hook_intelligence, get_image_ad_scan_preset, get_outfit_preset, get_powersource, get_saved_asset, get_saved_assets_batch, get_saved_visual_idea_preset, get_strategy, get_talent_model_preset, get_visual_preset_preset, get_visual_style_preset, list_ad_formula_presets, list_brand_assets, list_brand_documents, list_brands, list_cd_card_bookmark_presets, list_creative_agent_presets, list_creative_agent_skill_presets, list_creative_director_playbook_presets, list_decoded_ad_presets, list_image_ad_scan_presets, list_outfit_presets, list_projects, list_saved_assets, list_saved_visual_idea_presets, list_skills, list_strategies, list_strategy_audiences, list_strategy_tones, list_talent_model_presets, list_visual_preset_presets, list_visual_style_presets, load_skill, load_skill_reference, perplexity_search, read_brand_document, reclassify_brand_assets, retag_brand_asset, save_asset, search, search_community, search_research, search_skills
Every block below points at https://www.heista.co/api/mcp/mcp. Pick your client and copy.
claude mcp add --transport http heista https://www.heista.co/api/mcp/mcp
{
"mcpServers": {
"heista": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://www.heista.co/api/mcp/mcp"]
}
}
}{
"mcpServers": {
"heista": {
"url": "https://www.heista.co/api/mcp/mcp"
}
}
}{
"servers": {
"heista": {
"type": "http",
"url": "https://www.heista.co/api/mcp/mcp"
}
}
}{
"context_servers": {
"heista": {
"url": "https://www.heista.co/api/mcp/mcp"
}
}
}{
"mcpServers": {
"heista": {
"serverUrl": "https://www.heista.co/api/mcp/mcp"
}
}
}{
"mcpServers": {
"heista": {
"url": "https://www.heista.co/api/mcp/mcp"
}
}
}name: Heista
version: 0.0.1
schema: v1
mcpServers:
- name: Heista
type: streamable-http
url: https://www.heista.co/api/mcp/mcp{
"mcpServers": {
"heista": {
"httpUrl": "https://www.heista.co/api/mcp/mcp"
}
}
}{
"mcpServers": {
"heista": {
"type": "http",
"url": "https://www.heista.co/api/mcp/mcp"
}
}
}{
"mcpServers": {
"heista": {
"url": "https://www.heista.co/api/mcp/mcp"
}
}
}goose://extension?url=https%3A%2F%2Fwww.heista.co%2Fapi%2Fmcp%2Fmcp&type=streamable_http&id=heista&name=Heista
It answered our last check on 17 October 2026 in 281 ms. Over 10 checks since 4 September 2026 it answered 10. Status here comes from Neuronto's own probes, not from the publisher.
Yes. It is published by heista.co and served from www.heista.co, the vendor's own domain: https://www.heista.co/api/mcp/mcp.
118, as returned by the server's own tools/list on 9 September 2026: add_brand_asset, adformula_intelligence, call_creative_agent_preset, call_creative_worlds, chat_with_creative_worlds, check_balance, create_powersource_docs, create_powersource_full, and 110 more.
It answered our last check on 17 October 2026 in 281 ms. Over 10 checks since 4 September 2026 it answered 10.
Put the server under mcpServers in ~/.cursor/mcp.json with the URL https://www.heista.co/api/mcp/mcp. The exact JSON is above, with the same setup for Claude Code, Claude Desktop, VS Code, Windsurf, Zed, Gemini CLI and the rest.
It listed its tools to an anonymous client, so reading the tool list needs no key. Individual tools can still ask for credentials in their own inputs.
Publisher: co.heista. Endpoint host: www.heista.co. ARD identifier: urn:air:co.heista:mcp:api.
First indexed 31 August 2026. The tools are exactly what the server returned; nothing on this page
is inferred from its name. Ranking in Neuronto search is never for sale.