PitchSpark: Pre-Seed & Seed Deal Flow
Server Details
Search scored pre-seed & seed startups that are actively raising, by sector, stage or query.
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-11-25
- URL
- ChatGPT
- –
- Claude
- –
TDQS
Scored across 17 tools
The core intents (search, read pitch, score, decisions, compare, save/watch) are recognizable, but the surface contains three overlapping search versions and multiple read paths: get_pitch_v1, get_pitch_v2, get_score_v1, and get_decision_summary_v1 all return partially overlapping pitch/score/decision data. The descriptions give useful 'prefer X' guidance, but an agent still has meaningful misselection risk, especially between search v1/v2/v3 and get_pitch v1/v2.
Names are overwhelmingly snake_case with a pitchspark_ prefix and a verb_noun pattern (search_pitches, get_pitch, save_search, request_intro), plus consistent version suffixes. There are minor deviations such as pipeline_stage_v1 (noun phrase rather than verb_noun) and the presence of v1/v2/v3 parallel names, but the convention remains predictable.
At 17 tools, the server is slightly heavy but still within a reasonable range for a deal-flow product covering search, pitch reading, scoring, decisions, comparison, watchlist, saved searches, thesis fit, intros, and founder submission. The count is defensible, though version duplication inflates it and a few legacy variants could be collapsed.
The surface covers the main investor deal-flow lifecycle well: discovery via search, detailed pitch reading, scoring, decision aggregates, comparison, saved searches, watchlist, thesis fit, and intro requests, plus founder draft submission and health checking. Gaps remain around mutation of saved items and pipeline movement (pipeline_stage_v1 is read-only and explicitly cannot move deals), but these are minor relative to the core workflow.
Publisher details
- Operator
- Pitchspark · Publisher source
- Operator website
- https://www.pitchspark.io · Publisher source
- Vendor relationship
- Not applicable
- Documentation
- https://pitchspark.io/llms-full.txt · Publisher source
- Privacy policy
- https://pitchspark.io/PrivacyPolicy#ai-assistants · Publisher source
- Trust center
- Not applicable
- Restrictions
- Unknown
Available Tools
17 toolspitchspark_compare_pitches_v1Compare 2 to 4 startup pitches side by side: scores, traction, raise and decisionsARead-onlyIdempotentInspect
Compare 2 to 4 pitches side by side: ask, stage, sector, HQ region, traction note, green flags, public score, dimension scores (keyed callers, public scores only) and compact decision aggregates (likes vs passes, average rating, top pass reasons; investor API keys only, else reason key_required; each withheld below 5 distinct signed-in users, counts as bands; the likes-vs-passes block may be withheld with reason hidden_pending_review, and the team-related pass reasons are withheld pending review). Use when: an investor is choosing between shortlisted pitches. Do not use for: more than 4 pitches (run pitchspark_search_pitches_v2 sorted by score instead), full pitch text (pitchspark_get_pitch_v2). All ids must be distinct and resolvable: any id that is missing, not public or opted out answers NOT_FOUND for the whole call, and nothing is counted. The whole comparison is delivered or nothing is (a spent allowance answers RATE_LIMITED). Returns {"pitches":[...in the order given], "attribution"}. Each pitch this tool returns counts 1 against the daily pitch allowance (the same allowance as pitchspark_search_pitches_v1), or 0 when the same API key (without a key, the same connection) already got that pitch today. Calls are logged, and each pitch returned is counted as a view.
Inputs: ids (required). Returns: error, pitches, attribution
pitchspark_get_decision_summary_v1What signed-in PitchSpark users decided about a startup pitch: anonymised pass reasons, ratings and engagementARead-onlyIdempotentInspect
Summarise how signed-in PitchSpark users (not only investors) responded to one pitch: positive (likes and superlikes) vs passes (the thumbs block, currently withheld pending review: reason hidden_pending_review), the rating distribution and average (out of 10, to the nearest 0.5), the top pass reasons and "why we passed" themes (the team-related reasons and the team theme are withheld pending review and never listed; shares are over the listed reasons), time spent, and which pitch sections and score dimensions users read. Use when: an investor (or a founder's assistant) wants the crowd signal behind a pitch before deciding. Prefer it over pitchspark_get_pitch_v2 when only decision data is needed. Do not use for: who passed or liked (never returned), investor comments (never returned), the AI score (pitchspark_get_score_v1). Decision data from signed-in PitchSpark users needs a PitchSpark API key from an investor account (self-declared; PitchSpark does not verify accreditation): without one every decision block is available false with reason key_required. It is anonymised: counts of distinct people only, as bands ("5-9", "10-24", "25-49", "50+"), never who; shares and rates to the nearest 0.05; the pitch's founders, team and PitchSpark staff are never counted; guest and anonymous viewers are never counted; every block is withheld (reason below_min_k) when fewer than 5 distinct people are behind it, and a split (positive vs pass, rating bands) is withheld when any part of it would be under 5; no investor comment text is ever returned. Who is counted: ratings, reading time, sections and score views count ANY signed-in non-owner user (the writers check no account type), and likes, passes and pass reasons count signed-in users who swiped or gave pass feedback (also not restricted to investor accounts); only the caller of this tool must hold an investor key. Some fields are withheld pending a review (reason hidden_pending_review, or absent from a list): the team-related pass reasons and the pass rate (the likes-vs-passes thumbs block, the search card signal, investors_considered and the card-dwell figures). Treat a withheld field as no signal. Sources: signed-in users' swipe decisions (likes, passes, pass reasons, card dwell), structured pass feedback (never comment text), ratings, time on the pitch page and per-section dwell, and score-dimension views. A block with no source data is available false, reason no_data. Returns {"pitch_id","company_name","decisions":{min_k, investors_considered, truncated, thumbs, rating, pass_reasons, pass_themes, time_spent, sections, score_dimension_interest},"attribution"}. Without an investor API key the call still answers (and counts against the allowance like pitchspark_get_pitch_v1), but every block is withheld. Each pitch this tool returns counts 1 against the daily pitch allowance (the same allowance as pitchspark_search_pitches_v1), or 0 when the same API key (without a key, the same connection) already got that pitch today. Calls are logged, and each pitch returned is counted as a view.
Inputs: id (required). Returns: error, pitch_id, decisions, attribution, +1 more
pitchspark_get_health_v1Check that PitchSpark pitch search is upARead-onlyIdempotentInspect
Check that PitchSpark's pre-seed and seed deal-flow search is up before running a batch of startup searches. Returns {"status":"ok","version":"1.7.0"} and no pitch data. No key needed; uses none of the daily pitch allowance.
Inputs: no named parameters. Returns: error, status, version
pitchspark_get_pitch_v1Read one startup's full public pitch on PitchSpark, by id from a searchARead-onlyIdempotentInspect
Read the full public pitch of one early-stage startup from a PitchSpark search, by pitch_id: problem and solution, traction, market size, business model, competitors, moat, highlights, team size, HQ region, the ask, green flags and pitchspark_score (PitchSpark's AI pitch score, 0-100, published only at 70 or above, otherwise null). Use it to screen a pre-seed or seed deal in depth before requesting an intro. Founder claims are self-reported; financials, decks and founder contact details are not included.
Inputs: pitch_id (required). Returns: error, pitch, attribution
pitchspark_get_pitch_v2Read one startup pitch with its score, roast and anonymised decisionsARead-onlyIdempotentInspect
Read one startup pitch by id and choose what comes back with "sections": "pitch" (the full public pitch, the same object as pitchspark_get_pitch_v1), "score" (score and roast, as pitchspark_get_score_v1), "decisions" (anonymised decision aggregates by signed-in users, as pitchspark_get_decision_summary_v1). Default: all three. Use when: an investor wants the whole picture of one shortlisted pitch in one call. Prefer the single-purpose tools when you only need one part. Do not use for: searching (pitchspark_search_pitches_v2), founder contact details, financials, decks or data rooms (never returned), investment advice. Anonymous callers can use it under the daily allowance; dimension scores and analyses need an API key, and the decisions section an investor API key. Decision data from signed-in PitchSpark users needs a PitchSpark API key from an investor account (self-declared; PitchSpark does not verify accreditation): without one every decision block is available false with reason key_required. It is anonymised: counts of distinct people only, as bands ("5-9", "10-24", "25-49", "50+"), never who; shares and rates to the nearest 0.05; the pitch's founders, team and PitchSpark staff are never counted; guest and anonymous viewers are never counted; every block is withheld (reason below_min_k) when fewer than 5 distinct people are behind it, and a split (positive vs pass, rating bands) is withheld when any part of it would be under 5; no investor comment text is ever returned. Who is counted: ratings, reading time, sections and score views count ANY signed-in non-owner user (the writers check no account type), and likes, passes and pass reasons count signed-in users who swiped or gave pass feedback (also not restricted to investor accounts); only the caller of this tool must hold an investor key. Some fields are withheld pending a review (reason hidden_pending_review, or absent from a list): the team-related pass reasons and the pass rate (the likes-vs-passes thumbs block, the search card signal, investors_considered and the card-dwell figures). Treat a withheld field as no signal. An id that does not exist, a pitch that is not public and a pitch whose founders opted out of AI agents all answer NOT_FOUND. Each pitch this tool returns counts 1 against the daily pitch allowance (the same allowance as pitchspark_search_pitches_v1), or 0 when the same API key (without a key, the same connection) already got that pitch today. Calls are logged, and each pitch returned is counted as a view.
Inputs: id (required), sections. Returns: error, pitch, score, sections, +2 more
pitchspark_get_score_v1Get a startup pitch's PitchSpark AI score, dimension scores and roast commentaryARead-onlyIdempotentInspect
Get PitchSpark's AI pitch score for one pitch: the composite (0-100), the six dimensions (market, problem, traction, moat, financials, team, each 0-100 with the model's analysis) and the Roast Room commentary when the pitch was roasted. Use when: an investor asks how PitchSpark rates a pitch, or where it is strong and weak. Prefer it over pitchspark_get_pitch_v2 when only the score is needed. Do not use for: scoring your own text (pitchspark_roast_text_v1, when enabled), what signed-in users decided (pitchspark_get_decision_summary_v1), investment advice. Limits of the data: scores are published only at 70 or above; below that pitchspark_score is null and the dimensions are withheld (dimensions_withheld "not_public"): null means no public score, not a low one. Dimensions and analyses need an API key (dimensions_withheld "key_required" without one). Roast text is filtered for personal attacks and contact details. Same access as pitchspark_get_pitch_v1: anonymous callers allowed under the daily allowance. Returns {"score":{pitch_id, company_name, pitchspark_score, scale, public_min, dimensions, dimensions_withheld, scoring_version, scored_at, roast}, "attribution"}. Each pitch this tool returns counts 1 against the daily pitch allowance (the same allowance as pitchspark_search_pitches_v1), or 0 when the same API key (without a key, the same connection) already got that pitch today. Calls are logged, and each pitch returned is counted as a view.
Inputs: id (required). Returns: error, score, attribution
pitchspark_list_saved_searches_v1List the startup searches saved under your API keyARead-onlyIdempotentInspect
List the searches saved with pitchspark_save_search_v1 under the caller's API key, newest first, each with the exact arguments to re-run it ("rerun": tool pitchspark_search_pitches_v2 and its arguments). Use when: the investor asks for their saved searches. Do not use for: running them (pass rerun.arguments to pitchspark_search_pitches_v2). Requires a PitchSpark API key (psk_...). Searches are per key, not per account. Read-only; uses none of the daily pitch allowance.
Inputs: no named parameters. Returns: error, items, max_per_key
pitchspark_list_watchlist_v1Investors: list the startup pitches saved to your PitchSpark watchlistARead-onlyIdempotentInspect
List the pitches the investor has saved on PitchSpark (in the app or with pitchspark_watch_pitch_v1), newest first: id, company name, stage, sector, when saved and the link. Use when: the investor asks what is on their watchlist or saved pitches. Do not use for: saving (pitchspark_watch_pitch_v1), searching. Requires a PitchSpark API key from an investor account (self-declared; PitchSpark does not verify accreditation). Pitches that are no longer public or whose founders opted out of AI agents are left out and counted in unavailable_count. Up to 50 items. Uses none of the daily pitch allowance. Read-only. Calls are logged.
Inputs: limit. Returns: error, items, truncated, unavailable_count
pitchspark_pipeline_stage_v1Investors: read where your pitches sit in your PitchSpark deal pipelineARead-onlyIdempotentInspect
Read the investor's own PitchSpark deal pipeline: for each pitch, its stage (sparked, deck_requested, in_conversation, due_diligence, invested), when it last moved and whether it is archived. Optionally one pitch_id. Use when: the investor asks where a deal stands. Do not use for: moving a deal (not available through this server; use PitchSpark), other investors' pipelines (never). Requires a PitchSpark API key from an investor account (self-declared; PitchSpark does not verify accreditation). Only the key owner's own rows are read. Notes and investment amounts are never returned (has_notes says whether there are notes). Pitches no longer public or opted out are left out (unavailable_count). Up to 50 items. Read-only; uses none of the daily pitch allowance.
Inputs: pitch_id, include_archived. Returns: error, items, truncated, unavailable_count
pitchspark_request_intro_v1Investors: ask a startup founder for an intro through PitchSparkAInspect
For investors who found a startup to invest in: ask the founders of one PitchSpark pitch (by pitch_id) to connect, with a short message, through PitchSpark's in-app contact flow. Founders reply inside PitchSpark; their contact details are never returned. Returns status (requested, already_requested or quota_exhausted), intro_id and next_steps. Needs a PitchSpark API key (psk_...) from an investor account (self-declared; PitchSpark does not verify accreditation), or where OAuth is enabled a token for one with the intros:write scope. PitchSpark does not verify investors or founders, does not recommend investments and takes no part in any resulting deal.
Inputs: message (required), pitch_id (required). Returns: error, status, intro_id, next_steps
pitchspark_save_search_v1Save a startup search (query and filters) to re-run laterAIdempotentInspect
Save a search (the query and filters of pitchspark_search_pitches_v2) under the caller's API key, to list and re-run later with pitchspark_list_saved_searches_v1. Use when: the investor asks to save or remember a search. Do not use for: alerts (not delivered yet: saving does not send anything), running the search (pitchspark_search_pitches_v2). Requires a PitchSpark API key with the searches:write scope (keys without a scope list hold it). Idempotent: the same normalised query and filters under the same key returns the existing saved search (status "already_saved") and writes nothing new. At most 20 saved searches per key (status "limit_reached"); 50 saves per key per UTC day (RATE_LIMITED). Do not put personal data (names, emails, phone numbers, anything about a person) in "name" or "query": they are stored for up to 180 days. Emails and phone numbers that do appear are stored as [redacted]. Saved searches are deleted 180 days after they were last saved, and with the account. Returns {"status","saved_search","max_per_key","next_steps"}.
Inputs: name, sort, query, stage, +9 more. Returns: error, status, next_steps, max_per_key, +1 more
pitchspark_search_pitches_v1Find pre-seed and seed startup pitches for deal flow on PitchSparkARead-onlyInspect
Find early-stage startups to invest in: search scored pre-seed and seed pitches on PitchSpark in plain English ("pre-seed fintech in Europe under $2M") or by sector, stage, region, raise size, traction, green flags or score. Each Spark Card: ask, stage, sector, region, traction, green flags, pitchspark_score, pitch link. pitchspark_score is PitchSpark's AI pitch score (0-100, shown at 70+); founder claims are self-reported. Keyword matching, not semantic. No key needed for 50 pitches a day.
Inputs: query (required), limit, stage, sector, +8 more. Returns: error, pitches, truncated, attribution, +1 more
pitchspark_search_pitches_v2Search pre-seed, seed and early-stage startup pitches with filters, sorting and pagingARead-onlyInspect
Search founder-submitted startup pitches on PitchSpark with structured filters, five sort orders and cursor paging, returning lightweight cards with the public score, the top two score dimensions and a pass-rate signal (currently withheld pending review). Use when: building or paging through a filtered deal-flow list (stage, raise range, sector, HQ region, traction signals, score floor, recency), or sorting by score, recency, investor views or updates. Prefer it over pitchspark_search_pitches_v1 when you need paging, sorting or recency; pitchspark_search_pitches_v1 stays available unchanged. Do not use for: reading one pitch in depth (pitchspark_get_pitch_v2), score detail (pitchspark_get_score_v1), decision detail from signed-in users (pitchspark_get_decision_summary_v1), founder contact details, decks or data rooms (never returned), investment advice. How search works: rule-based keyword + synonym ranking over the founders' public text; there is NO semantic or embedding search yet, so concrete words work better than abstract ones. query is optional: filters alone list every match. Paging: limit (default 10; the ceiling is 10 without a key or with a free key and 25 with a paid key) and either cursor (from page.next_cursor) or offset; results stop at offset 200. Unlike pitchspark_search_pitches_v1, pitches seen before are not left out (paging needs a stable list). Returns "pitches" (each card: the pitchspark_search_pitches_v1 card keys, plus pitchspark_score (public only, 70+, else null), top_dimensions (the two strongest of the six score dimensions, 0-100, keyed callers only and only when the score is public, else null), decision_signal ({available, reason, decisions_band, pass_rate, min_k}: likes vs passes by signed-in users from stored counters, for investor API keys only (reason key_required otherwise), withheld unless there are at least 5 decisions and the likes and the passes are each 0 or at least 5; the pass rate is currently hidden pending review: reason hidden_pending_review, treat it as no signal), created_at and updated_at), "page" ({offset, limit, returned, total_matches, next_cursor, max_offset}), "sort", "search_mode" ("keyword_synonym"), "interpretation" and "attribution". Repeated identical queries are served from a short-lived cache. Each pitch this tool returns counts 1 against the daily pitch allowance (the same allowance as pitchspark_search_pitches_v1), or 0 when the same API key (without a key, the same connection) already got that pitch today. Calls are logged, and each pitch returned is counted as a view. Recommended flow: pitchspark_search_pitches_v2 (find) → pitchspark_get_pitch_v2 (read) → pitchspark_get_score_v1 (score and roast) → pitchspark_get_decision_summary_v1 (what signed-in PitchSpark users decided) → pitchspark_request_intro_v1 or pitchspark_watch_pitch_v1. Errors are JSON {"error":{"code","message","category","retryable"}}: INVALID_ARGUMENT, UNAUTHORIZED, RATE_LIMITED, INTERNAL.
Inputs: sort, limit, query, stage, +11 more. Returns: page, sort, error, pitches, +4 more
pitchspark_search_pitches_v3Search pre-seed and seed startup pitches by meaningARead-onlyInspect
Find pre-seed and seed startups that are actively raising by what the investor means, not only the words used (e.g. "software that helps independent clinics get paid faster" finds revenue-cycle and payments pitches that never say "get paid"). Use when: an investor describes the kind of company they want in their own words (a problem, a customer, a business model) rather than exact keywords. Prefer it over pitchspark_search_pitches_v2 for descriptive queries; use pitchspark_search_pitches_v2 to sort by score, recency or views. Do not use for: reading one pitch in depth (pitchspark_get_pitch_v2), founder contact details, decks or data rooms (never returned), investment advice. The filters (sector, stage incl. "early", HQ region, raise range, minimum public score, traction, green flags, listed or updated within N days) narrow the list first; the query then ranks what is left. Ranking uses embeddings when they are available (similarity blended with the public score and recency) and falls back to the keyword + synonym ranker of pitchspark_search_pitches_v2 when they are not. Every reply states which in "search_mode" (semantic, mixed or keyword_fallback) with "fallback_reason". Returns the same lightweight cards as pitchspark_search_pitches_v2, "page" (cursor paging) and "coverage". Email addresses and phone numbers are removed from the query before it is used. Each pitch this tool returns counts 1 against the daily pitch allowance (the same allowance as pitchspark_search_pitches_v1), or 0 when the same API key (without a key, the same connection) already got that pitch today. Calls are logged, and each pitch returned is counted as a view. Errors are JSON {"error":{"code","message","category","retryable"}}: INVALID_ARGUMENT, UNAUTHORIZED, RATE_LIMITED, INTERNAL.
Inputs: query (required), limit, stage, cursor, +9 more. Returns: page, error, pitches, coverage, +4 more
pitchspark_submit_pitch_v1Founders only: create a draft of your own startup pitch on PitchSparkAInspect
For founders raising a pre-seed or seed round: create a draft pitch on your own PitchSpark account from five facts (company name, sector, stage, raise amount in USD, a one-paragraph problem statement). The draft is not listed, searchable or scored until the founder completes it and submits it for review in Pitch Studio at the returned edit_url. Needs a founder submit token (pss_..., from PitchSpark Settings). Where OAuth is enabled, a token with the pitches:submit scope also works. Returns pitch_id, status "draft", company_name and edit_url.
Inputs: stage (required), sector (required), company_name (required), raise_amount_usd (required), +1 more. Returns: error, status, edit_url, pitch_id, +1 more
pitchspark_thesis_fit_v1Rank startup pitches against an investment thesis, with a fit rationaleARead-onlyInspect
Rank PitchSpark pitches by how well they fit an investment thesis: either thesis text you pass, or the investor's saved thesis and preferences on PitchSpark (use_saved_thesis). Each result has a fit_score (0-100) and a rationale listing exactly which criteria matched. Use when: an investor asks "what fits my thesis". Prefer pitchspark_search_pitches_v2 for explicit filters. Do not use for: investment advice; the fit is a rule-based match, not a judgement. How it works (rule_based_v1, no model call): the thesis is read for sectors, stages, HQ regions and keywords (the same rules as search); a pitch earns points for each criterion it meets (sector 35, stage 25, region 10, keywords up to 30), scaled to the criteria the thesis names. Pitches that meet none are left out. Requires a PitchSpark API key (psk_...). The thesis text is never logged or stored. Returns {"pitches":[{card, fit_score, rationale}], "thesis", "method", "attribution"}. Up to 10 results (25 on a paid key). Each pitch this tool returns counts 1 against the daily pitch allowance (the same allowance as pitchspark_search_pitches_v1), or 0 when the same API key (without a key, the same connection) already got that pitch today. Calls are logged, and each pitch returned is counted as a view.
Inputs: limit, thesis_text, use_saved_thesis. Returns: error, method, thesis, pitches, +2 more
pitchspark_watch_pitch_v1Investors: save a startup pitch to your PitchSpark saved pitchesAIdempotentInspect
For investors building a deal-flow shortlist: save one startup pitch (by pitch_id) to your PitchSpark saved pitches, to come back to on pitchspark.io. The founders are not contacted; the save counts as interest in their pitch statistics, where investors are numbered and never named. Saving again changes nothing. Returns status (watching, already_watching or limit_reached), pitch_id and next_steps. Needs a PitchSpark API key (psk_...) from an investor account (self-declared; PitchSpark does not verify accreditation), or where OAuth is enabled a token for one with the watchlist:write scope.
Inputs: pitch_id (required). Returns: error, status, pitch_id, next_steps
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
Search startups, look up scores, request research memos and read public startup research.
Search institutional pre-seed venture capital firms, partner profiles, and valuation.
Search 3,650+ institutional venture capital firms and 13,000+ partners by stage, sector, and HQ.
Live index of startups that just raised funding, verified daily. Recent raises, lookups, stats.
Related MCP Servers
AlicenseNot gradedqualityDmaintenanceProvides VC-grade startup intelligence, allowing founders to validate ideas and VCs to screen deals using tools like scoring, investor matching, and financial analysis.9 npmMIT- AlicenseNot gradedqualityBmaintenanceEnables searching and inspecting open roles at startups that closed a venture round in the last six months, including full job details, application form questions, and company funding and hiring data. Also supports a personal workspace for tracking applications, curating company lists, and saving searches with email alerts.MIT
- AlicenseNot gradedqualityCmaintenanceEnables agents to research venture capital and angel investors for a fundraise — searching firms by stage, sector and geography, seeing which investors backed which companies and rounds, and identifying the specific partner to contact. Every returned fact carries the source page it came from, with contact routes limited to publicly published ones and a read-time removal list honoring opt-outs.78 npmMIT
- AlicenseNot gradedqualityDmaintenanceEnables detection and analysis of pre-public product launches through web search, content extraction, AI-powered scoring, and automated alerting. Provides comprehensive tools for surfacing stealth startup signals before they trend publicly.MIT
Glama MCP Gateway
Add one secure layer between your agents and this server.