# Fund Momentum MCP — tools reference

Plain Markdown on purpose. Readable by a person, and by whatever assistant you
paste this URL into.

- **Endpoint:** `https://fundmomentum.vc/_api/mcp`
- **Transport:** Streamable HTTP (MCP)
- **Coverage:** 1,175 actively deploying VC funds, and 814 disclosed LPs across 65 HQ countries, 200 of which have backed emerging managers
- **LP data, two tools:** `check_lp_coverage` is free forever, for every caller,
  and returns COUNTS ONLY — how many LPs match a country and LP type — with no
  key needed. `search_lps` returns full LP records and is billed exactly like
  `get_fund`: free on the keyless trial (the same 10-per-day pool as every
  keyless call), free within the monthly "profile pull" allowance it shares with
  `get_fund` (Free 20, Starter 200, one pull per LP row), unlimited in Pro, and billed
  €0.01 per LP row on agent credits. LP Radar is included in Pro (no separate subscription). search_lps is billed EUR 0.01 per LP row on agent credits. There is no separate LP
  subscription gating this tool any more. Commitment history is not served over MCP at all.
- **10 of 11 tools are read-only** and carry `readOnlyHint: true`: nothing
  you call there changes anything, in your account or in ours. The exception is
  `send_feedback` (`readOnlyHint: false`), which stores one short message and
  touches no fund, account or billing data.
- **`get_usage` and `send_feedback` are free, need no key and do not count against
  the keyless 10 calls per day.** `get_usage` reports your tier, calls used and remaining,
  credits (agent keys), reset time and per-minute pace limit, plus profile pulls
  (`pulls_used`, `pulls_limit`, `pulls_remaining`, `pulls_resets_at`). `send_feedback` takes a
  `message` (max 1000 characters; leave out personal details of third parties),
  an optional `category` and `tool_name`, and is limited to 5 per caller per UTC day.
- **A keyless caller with 2 or fewer calls left today** gets one extra short text item
  after the tool's JSON result, with the calls left, the reset time and the pricing
  link. The JSON payload and `_meta` are not changed.
- **`search_funds` returns 10 funds when `limit` is omitted, for every caller**
  (changed 2026-10-25: paying agents and Starter/Pro keys used to receive every match, or the
  per-call ceiling). Pass a larger `limit` explicitly; the per-caller caps are unchanged.

## Try it with no key at all

```bash
curl -s -X POST https://fundmomentum.vc/_api/mcp \
  -H 'Content-Type: application/json' \
  -H 'Accept: application/json, text/event-stream' \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call",
       "params":{"name":"search_funds","arguments":{"country":"Austria","limit":3}}}'
```

`search_funds`, `get_fund`, `get_changes`, `check_lp_coverage`, `search_lps` answer 10 calls per caller per
UTC day, combined, with no credential and no signup.

## Add it to a client

```bash
claude mcp add --transport http fund-momentum https://fundmomentum.vc/_api/mcp
```

Cursor, Windsurf, VS Code and anything else that speaks remote MCP over
Streamable HTTP take the same URL. ChatGPT's connectors cannot send a custom
header, so pass the key in the URL instead: `https://fundmomentum.vc/_api/mcp?api_key=YOUR_KEY`
with "No authentication" selected.

## The order to call things in

Most failed calls against this server are a slug that was invented rather than
looked up. The sequence that works:

1. **`search_funds`** — or `match_startup` if the brief is qualitative. Both
   return a `slug` for every result.
2. **`get_fund`** — pass a slug from step 1, never one built from a fund's name.
   A near-miss is resolved only when exactly one fund matches (different case or
   spacing, or a name that is a hyphen-boundary prefix of one real slug, such as
   `speedinvest-africa`). It is then served with `resolved_from` set to what you
   asked for; use the real slug next time. A name matching several funds
   (`speedinvest`) is listed, not guessed. Applies to `get_fund`,
   `get_fund_signals` and `get_gp_profile`, before any payment.
3. **`get_fund_signals`** or **`get_gp_profile`** — drill into a fund you have
   already identified.
4. **`get_changes`** — afterwards, to stay current without re-running step 1.

For limited partners: **`check_lp_coverage`** first (free, counts only, no key
needed) to see whether the geography and LP type you need are covered, then
**`search_lps`** for the records — priced exactly like `get_fund` (see Tiers).

## Free tools

### `search_funds`

**Search VC funds** · Free tier · read-only

START HERE for any question about funds. Search actively deploying venture capital funds by stage, country, industry and recent closes. Works with no API key at all (keyless trial); see the Cost line at the end for per-fund pricing on agent credits. Returns name, slug, country, stage, fund size, industries, the latest published close (lastCloseDate, lastCloseType) and a durable profile URL. For 'which funds just closed', pass closed_since: results then come newest close first. The `slug` in each result is what every other tool takes as input, so run this first and copy the slug rather than constructing one from a fund's name.
Cost: 1 credit per fund returned (EUR 0.01 each). Default limit 10. Default max_credits 10, hard maximum 25 per call. Billed per fund for agent credits and inline MPP only; Starter and Pro are flat-fee and the keyless trial is free. The cost of each call is reported in the response's billing object.

| Parameter | Type | Required | Notes |
|---|---|---|---|
| `stage` | string | no | Funding stage. Matches every fund that invests at this stage, including multi-stage funds whose range covers it (each result lists all its stages in fundingStages). Must be one of the listed enum values (lowercase, underscores). Common spellings such as 'Pre-Seed' or 'Series A' are accepted and normalised. One of 7 enum values. |
| `country` | string | no | Country of the fund's headquarters, spelled out in full (e.g. 'United States', 'United Kingdom', 'Germany'). Not an ISO code. |
| `industry` | string | no | Industry focus. Must be one of the listed enum values (lowercase, underscores), e.g. 'ai_ml', 'fintech', 'climate_sustainability'. Common spellings such as 'AI/ML', 'FinTech' or 'climate' are accepted and normalised. One of 54 enum values. |
| `closed_since` | string | no | Only funds whose latest published close (first, interim or final close, or a launch) is on or after this date, ISO format YYYY-MM-DD (e.g. '2026-09-01'). Close data comes from Fund Momentum's own close coverage, each with a source URL in get_fund's last_close block. |
| `limit` | integer | no | Number of results to return. Default 10 for every caller, including agents paying with credits and subscribers, so an omitted limit never returns or bills more than 10 funds. Without a paid account the maximum is 10; a larger value is rejected, not clamped. Agents paying with credits are billed 1 credit (EUR 0.01) per fund returned, up to 150 funds per key per UTC day (a call that would exceed the remainder is refused before any charge); pass a larger limit explicitly to receive more. Starter gets up to 50 per call; Pro gets up to 250 and can page with offset. The response reports `matched` (total matches) and `truncated`. |
| `offset` | integer | no | Number of matches to skip, for paging through a large result set (use next_offset from a truncated response). Agents paying with credits and Pro plans only. |
| `max_credits` | integer | no | Optional spending ceiling for this call, in credits (1 credit = EUR 0.01). Rows returned never exceed floor(max_credits / credits per row). Omitted: the ceiling is the cost of the effective limit (10 credits for the default page). Values above 25 are clamped to 25; zero, negative and non-integer values are rejected before anything is charged. It is a clamp on limit only and does not apply to Starter or Pro. When it cuts the page, the response has truncated: true and truncated_reason: "max_credits". |

**Accepted `stage` values** (7):

```
growth, late_stage, pre_seed, seed, series_a, series_b, series_c
```

**Accepted `industry` values** (54):

```
agritech, ai_ml, biotech, built_environment, carbon_removal, circular_economy, climate_sustainability, cloud_devops, consumer_commerce, creator_economy, cybersecurity, data_infrastructure, deep_tech, defense, developer_tools, diagnostics, digital_banking, digital_education_consumer, digital_health, digital_infrastructure, ecommerce, energy_tech, enterprise_software, evs_charging, fashion_tech, fem_tech, fintech, food_beverage, food_tech, gaming, infrastructure_tech, insur_tech, iot, manufacturing, marketplaces, med_tech, mental_health, mobility_transport, payments_embedded_finance, private_markets, prop_tech, robotics_automation, saas, semiconductors, space, supply_chain, therapeutics, travel_tech, vertical_saas, water_tech, wealth_tech, web3_finance, web3_infrastructure, wellness_lifestyle
```


### `get_fund`

**Get fund profile** · Free tier · read-only

THEN: get the detailed profile of one fund, using a slug returned by search_funds. COST: free, no API key required. Every response carries a `provenance` block stating where each field came from (including per-field sources), when it was last verified, and how confident Fund Momentum is in it, plus a `last_close` block (close type, date, amount and currency as stated, source URL) when Fund Momentum has covered a close of this fund. Also carries `michael_network: {has_warm_path: boolean}` on EVERY caller, free, unconditional: whether Michael Schneider, Founder of Fund Momentum, may have a direct path to this fund personally. When true: a Pro caller also gets the matched contact's name; everyone else gets a teaser (no content, just confirmation it exists) pointing at request_warm_intro or the fund's own page -- credits, not a subscription, unlock it. Do not guess the slug from a fund's display name; it will usually be wrong and the call is wasted. A near-miss is resolved only when exactly one fund matches, and the result then carries `resolved_from`; a name matching several funds is listed, not guessed.
Cost: free with no key (10 keyless calls per day) or within a plan quota; 1 credit (EUR 0.01) per call on an agent key.

| Parameter | Type | Required | Notes |
|---|---|---|---|
| `slug` | string | yes | The slug identifier of a fund, exactly as returned in the `slug` field of search_funds. Call search_funds first rather than guessing: slugs are not derivable from a fund's display name. |


### `get_changes`

**Get changed funds since** · Free tier · read-only

FOR STAYING CURRENT, not for discovery: return only the funds whose data changed since a given timestamp. COST: free, no API key required. With no key it reaches back at most 7 days; a free key widens that to 30 days, and a paid caller is not limited. A shortened window is declared as since_floor_applied, never silently truncated. Designed for cheap repeated polling: send the `next_since` AND the `etag` from your previous response, and an unchanged window answers with `unchanged: true` and no rows. Each row has change_type created, signal_updated or close_updated; close_updated means Fund Momentum published a new first, interim or final close for that fund and carries a last_close block with type, date, size and source URL. Use this instead of re-running search_funds on a schedule. One call here replaces a full re-crawl. Polls that return no changes are free: they do not count against quotas or credits (the per-minute pace limit still applies).
Cost: free with no key (10 keyless calls per day) or within a plan quota; 1 credit (EUR 0.01) per call on an agent key. Empty polls are free.

| Parameter | Type | Required | Notes |
|---|---|---|---|
| `since` | string | no | Lower bound, exclusive. Accepts either an ISO 8601 timestamp ('2026-08-01T00:00:00Z') or a unix epoch in milliseconds as a string ('1785955265103'). Omit to receive the last 7 days. |
| `if_none_match` | string | no | The `etag` value from your previous get_changes response. If nothing changed, the reply is a few bytes instead of a full page. This is the cheapest poll available here. |
| `limit` | integer | no | Maximum rows to return (1-200) |


### `search_lps`

**Search LP records** · Free tier · read-only

THEN: return limited partner records (name, slug, LP type, HQ country and city, geographic focus, website, emerging-manager backing, verified flag, and michael_network: {has_warm_path: boolean, matches?}) filtered by country, LP type and emerging-manager backing, at most 25 per call. has_warm_path is free and unconditional on every caller; when true, a Pro/Enterprise caller also gets the matched contact's name in matches, everyone else gets a teaser (no content, just confirmation it exists) pointing at request_warm_intro or the LP's own page -- credits, not a subscription, unlock it. This is Michael Schneider's own personal network, not a data-matching feature; no data source is ever named. Set backs_emerging_managers=true to see only LPs that have already backed a first or second fund — for a first-time manager that is usually the only filter that matters. Billed exactly like get_fund, per caller: no key, 10 free calls per UTC day (the same pool as every keyless tool); a free/Starter/Pro key draws from a shared monthly "profile pull" allowance with get_fund, and search_lps counts one pull per LP row returned (Free 20/mo, Starter 200/mo, Pro unlimited with a fair-use limit of 500 LP rows per UTC day per key, error_reason fair_use_daily_lp_rows). If the allowance runs short the page is cut to what remains (truncated_reason "pull_quota") instead of failing; at zero you get pull_quota_exceeded with pulls_used, pulls_limit and upgrade_url. LP Radar is included in Pro, there is no separate subscription. An agent key is billed per LP returned at the same per-row price as search_funds (EUR 0.01 per row), with no monthly cap since every row is paid. Use check_lp_coverage first — it is free and unmetered, and tells you whether the dataset covers your geography and LP type before you call this. website is null when no website is on record; it is never omitted.
Cost: free with no key (10 keyless calls per day, one shared pool); a free/Starter/Pro key draws from the same monthly "profile pull" allowance as get_fund, one pull per LP row (Free 20/mo, Starter 200/mo, Pro unlimited with a fair-use limit of 500 LP rows per UTC day); LP Radar is included in Pro. An agent key is billed per LP returned at the same per-row price as search_funds (1 credit, EUR 0.01 each), with no monthly cap since every row is paid.

| Parameter | Type | Required | Notes |
|---|---|---|---|
| `country` | string | no | Headquarters country of the LP, spelled out in full (e.g. 'Germany'). Two-letter ISO codes are accepted too. |
| `lp_type` | string | no | LP type. Must be one of the listed values; common spellings such as 'family office' or 'pension' are accepted and normalised. One of 15 enum values. |
| `backs_emerging_managers` | boolean | no | true returns only LPs that have backed an emerging manager; false returns only those that have not. Omit for both. Not available on check_lp_coverage, which reports the figure as a breakdown instead. |
| `limit` | integer | no | Records to return, 1-25. A value above 25 is rejected, not clamped. |

**Accepted `lp_type` values** (15):

```
Pension fund, Insurer, Foundation / Endowment, University / Research, Bank, Public / State institution, Sovereign wealth fund, Fund-of-Funds, Asset manager / Investment firm, Family office / Holding, Corporate / Strategic, Impact investor, Investment vehicle, Individual / Angel, Institutional LP
```


### `get_usage`

**Get my usage and limits** · Free tier · read-only

START HERE for 'how many calls or credits do I have left'. COST: free, no API key required, and it does not count against the keyless 10 calls per day (it is subject to the per-minute pace limit like every call). Returns your tier (keyless, free, starter, pro, enterprise or agent), calls used and remaining for the current window (daily for keyless, monthly for plans), profile pulls (pulls_used, pulls_limit, pulls_remaining and when that window resets, pulls_resets_at; a pull is one get_fund call or one LP row from search_lps), credits remaining for agent keys, when the call window resets (resets_at), and your per-minute pace limit. Never charges anything.
Cost: free

_No parameters._


### `send_feedback`

**Send feedback to Fund Momentum** · Free tier · writes one feedback message

Tell the Fund Momentum team about a bug, a data problem or a missing feature. COST: free, no API key required, and it does not count against the keyless 10 calls per day; limited to 5 per caller per UTC day. Stores one short message and nothing else happens: nobody is notified instantly and there is no reply. Describe the problem generically and leave out names, email addresses and other personal details of third parties (founders, GPs, LPs); never paste private documents or credentials.
Cost: free

| Parameter | Type | Required | Notes |
|---|---|---|---|
| `message` | string | yes | What went wrong or what is missing, in plain words. Maximum 1000 characters. No personal details of third parties. |
| `category` | string | no | What kind of feedback this is. Defaults to other. One of 4 enum values. |
| `tool_name` | string | no | The tool the feedback is about, if any, exactly as listed in tools/list. |

**Accepted `category` values** (4):

```
bug, data_issue, missing_feature, other
```


## Paid tools

Available on a Pro key, on agent credits, or per call over MPP. A request for a
fund that does not exist, or one with no published signals, returns
`not_found` **before** any payment is taken.

**Exception: `request_warm_intro`.** Every other tool below is INCLUDED in
Pro -- a Pro key calls it for free, inside the monthly quota. `request_warm_intro`
is not: no tier, Pro or otherwise, ever makes an intro free. A Pro key pays from
its own per-type intro-credit balance (the same one the website's "Request a
warm intro" button spends), and an agent key pays per call at the identical
flat price. Visibility -- `get_fund`'s and `search_lps`' `michael_network`
field, `check_lp_coverage`'s `known_warm_path` -- is free on every tier; the
intro itself never is. An intro can only be requested where a known path exists
(`has_warm_path` / `michael_network`). For any other target the call creates a
free path check instead (`status: "pending_review"`, `path_check: true`,
`charged: false`): Michael checks his network by hand, the account holder decides
on the website whether to spend a credit only if a path is found, and an
unanswered check closes after 14 days. Optional `contact` and `context`
parameters apply. At most 5 open path checks per account (429,
`error_reason: "path_check_limit"`).

### `get_fund_signals`

**Get fund investor signals** · pro tier (€19/mo) · read-only

DRILL DOWN on a fund you already identified with search_funds or get_fund. Returns investor signals: GP thesis, bullish and contrarian signals, founder dos and don'ts, deployment status and source URLs. This is the tool to use before writing an outreach message to a specific fund. COST: included in Pro, or €0.10 per call over MPP if you have no key. Roughly 96% of funds carry signals; the rest return not_found BEFORE any payment is taken, so a wrong slug is never charged for.
Cost: included in Pro; otherwise EUR 0.10 per call (10 credits) on an agent key, or the same amount paid inline over MPP with no key.

| Parameter | Type | Required | Notes |
|---|---|---|---|
| `slug` | string | yes | The fund's slug, exactly as returned in the `slug` field of search_funds. Call search_funds first rather than guessing. |


### `get_gp_profile`

**Get General Partner profiles** · pro tier (€19/mo) · read-only

DRILL DOWN to the people: General Partner profiles for a fund you already identified. Use this when the question is who to approach and what they personally focus on, rather than what the fund does. COST: included in Pro, or €0.10 per call over MPP if you have no key. A fund with no published GP data returns not_found before any payment. Requires a slug from search_funds.
Cost: included in Pro; otherwise EUR 0.10 per call (10 credits) on an agent key, or the same amount paid inline over MPP with no key.

| Parameter | Type | Required | Notes |
|---|---|---|---|
| `fund_slug` | string | yes | The fund's slug, exactly as returned in the `slug` field of search_funds. Call search_funds first rather than guessing. |
| `gp_name` | string | no | Optional GP name to filter by |


### `match_startup`

**Match startup to funds** · pro tier (€19/mo) · read-only

ALTERNATIVE ENTRY POINT when you do not know what to search for. Describe a startup in plain language and receive the top 10 matching funds, each with a slug, a reason and a score. Use this instead of search_funds when the brief is qualitative ('climate hardware, pre-seed, Europe') rather than a filter. Slower than the other tools, three to ten seconds, because it reasons over every tracked fund: allow for that before you time out and retry. COST: included in Pro, or €0.25 per call over MPP if you have no key.
Cost: included in Pro; otherwise EUR 0.25 per call (25 credits) on an agent key, or the same amount paid inline over MPP with no key.

| Parameter | Type | Required | Notes |
|---|---|---|---|
| `description` | string | yes | Startup description (max 500 chars) |
| `stage` | string | no | Funding stage. Must be one of the listed enum values; common spellings are normalised. One of 7 enum values. |
| `country` | string | no | Startup HQ country, spelled out in full. A soft boost for funds headquartered there, not a hard filter. |
| `region` | string | no | Optional region. A soft boost for funds headquartered in, or declaring a focus on, that region; never excludes a fund. One of 5 enum values. |

**Accepted `stage` values** (7):

```
growth, late_stage, pre_seed, seed, series_a, series_b, series_c
```

**Accepted `region` values** (5):

```
DACH, EU, Nordics, UK, US
```


### `request_warm_intro`

**Request a warm intro** · pro tier (€19/mo) · writes one feedback message

Ask Michael for a warm intro to a fund or an LP. Fulfilled BY HAND through the same queue and Slack notification the website's own "Request a warm intro" button uses -- this is not an automated connection and carries no guaranteed timeline. A known match (get_fund's or search_lps' michael_network.has_warm_path) is a PRECONDITION: intros are only sold where Michael's network already has a path, exactly like the website. For a target without one nothing is sold: the call creates a FREE path check instead (status "pending_review", charged: false). Michael then checks his own network by hand; only if he finds a path does the account holder decide on https://fundmomentum.vc/intro-credits whether to spend one credit. No path means no charge, and an unanswered check closes automatically after 14 days. Asking again about a target you already asked about is free and returns the existing status; only a genuinely new request is charged. COST: EUR 50 per intro, flat -- the identical per-unit price a human pays on the intro-credits page: pay exactly quantity times EUR 50. No agent discount, and NO subscription tier (human or Pro) ever makes an intro free -- Free/Starter/Pro/LP Radar unlock visibility only (get_fund, check_lp_coverage, search_lps); an actual intro always costs, on a per-user/per-type credit balance for a human's own key, or on agent credits/MPP at this same price for an agent key. IDENTITY REQUIRED: unlike every other tool here, this one needs someone to contact, because Michael reviews and fulfills every request by hand. A human's own API key works directly. An agent key must first be linked to a human account (POST to /_api/agent/link; see the server card) -- an unlinked agent key is refused here before anything is charged.
Cost: EUR 50.00 per intro, flat. Never included in any plan (Free, Starter, Pro or Enterprise); a human key spends its intro-credit balance, an agent key pays per call at the same price. Re-asking about a target already requested is free.

| Parameter | Type | Required | Notes |
|---|---|---|---|
| `target_type` | string | yes | Whether `slug` identifies a limited partner or a fund. One of 2 enum values. |
| `slug` | string | yes | The slug of the LP (from search_lps or check_lp_coverage's referral) or the fund (from search_funds, match_startup or get_fund), exactly as returned by that tool. Not guessed from a display name. |
| `contact` | string | no | Optional. The named person at the target you want to be introduced to (max 80 characters), e.g. "Jane Doe". |
| `context` | string | no | Optional. Short note for Michael: what is being raised, how you are connected, timing (max 1000 characters). Saved with the request and shown to Michael, whether it becomes a paid intro or a free path check. |

**Accepted `target_type` values** (2):

```
lp, fund
```


## check_lp_coverage

The one tool that is free forever, for every caller, on any key or none, and
never metered or billed: counts only, never the records themselves. Use it to
see whether the dataset covers a geography and LP type before spending a pull
or a credit on `search_lps`, which returns at most 25 records per call.

### `check_lp_coverage`

**Check LP coverage** · Free tier · read-only

START HERE for any question about limited partners (LPs). Answers "does Fund Momentum cover my geography and LP type" with COUNTS ONLY: how many disclosed LPs match, how many of those have backed emerging managers, and how many have a known warm path in Michael's own network (`known_warm_path`). It never returns a name, slug, website or commitment. COST: free, works with no API key at all (counts against the keyless daily allowance), and is never billed to agent credits or a monthly quota. Counts below 5 are reported as "<5" and zero as "none". Call this before search_lps to find out whether LP Radar covers what you need.
Cost: free. Never billed to agent credits or a monthly quota; counts against the keyless allowance of 10 calls per day.

| Parameter | Type | Required | Notes |
|---|---|---|---|
| `country` | string | no | Headquarters country of the LP, spelled out in full (e.g. 'Germany', 'United States'). Two-letter ISO codes are accepted too. LPs whose headquarters is undisclosed are never counted under a country; they are reported separately as undisclosed_hq_count. |
| `lp_type` | string | no | LP type. Must be one of the listed values; common spellings such as 'family office', 'pension' or 'fund of funds' are accepted and normalised. One of 15 enum values. |

**Accepted `lp_type` values** (15):

```
Pension fund, Insurer, Foundation / Endowment, University / Research, Bank, Public / State institution, Sovereign wealth fund, Fund-of-Funds, Asset manager / Investment firm, Family office / Holding, Corporate / Strategic, Impact investor, Investment vehicle, Individual / Angel, Institutional LP
```


`search_lps` itself is listed above with the other free/paid tools — as of
2026-11-03 it is billed exactly like `get_fund`, not sold as a separate
subscription.

## Tiers

| Tier | Price | Calls | Profile pulls / month | Pace limit | search_funds max per call | Tools |
|---|---|---|---|---|---|---|
| Keyless trial | €0 | 10 per UTC day | shares the daily pool | 120 per minute per address | 10 | search_funds, get_fund, get_changes, check_lp_coverage, search_lps |
| Free (key) | €0 | 300 per month | 20 | 20 per minute | 10 | search_funds, get_fund, get_changes, search_lps, get_usage, send_feedback |
| Starter | €9/mo | 1,000 per month | 200 | 30 per minute | 50 | free tools |
| Pro | €19/mo | 10,000 per month | unlimited (fair use: 500 LP rows per UTC day per key) | 60 per minute | 250 | all 9 other fund tools free, including `search_lps` (LP Radar is included in Pro); `request_warm_intro` always costs a credit, see above |
| Enterprise | Custom | unlimited | unlimited | 120 per minute | 250 | all tools |
| Agent | €0.01/call | metered on credits | none (every pull is paid per row; `search_lps` €0.01 per LP row) | 60 per minute and 2,000 per UTC day | 10 unpaid default; paid cap 150 funds per key per UTC day | All 11 tools work on agent credits, except the free ones that need no credits (check_lp_coverage, get_usage, send_feedback). request_warm_intro also needs the agent key to be linked to a human account. |

"Profile pulls" is a second, narrower counter shared by `get_fund` (1 pull per
call) and `search_lps` (1 pull per LP record returned) — it sits inside the
general call quota, not beside it. `search_funds` rows are not pulls.
`check_lp_coverage` never touches either counter and is free on every tier.
When a Free, Starter or keyless caller's allowance runs short, `search_lps`
returns only the rows that remain and sets `truncated: true` with
`truncated_reason: "pull_quota"`; at zero it answers `pull_quota_exceeded`.
`get_usage` and `GET /_api/v1/usage` report `pulls_used`, `pulls_limit` and
`pulls_resets_at`.

Agents can self-register and receive a key plus 25 free credits immediately:

```bash
curl -s -X POST https://fundmomentum.vc/_api/agent/register \
  -H 'Content-Type: application/json' \
  -d '{"agent_name":"your-agent","email":"you@example.com"}'
```

## Errors worth knowing

| Code / status | Meaning |
|---|---|
| `-32042` | Payment required. A challenge is attached in `error.data.challenges`. Nothing has been charged |
| `-32043` | A payment credential failed verification. Money may have moved — report it and we will settle it by hand |
| `-32000`, `error_reason: "not_found"`, `checked_before_payment: true` | The slug does not exist, matches several funds (none is picked for you), or has no signals. You were not charged, and the message names the closest real slugs. A single unambiguous near-match is served instead, with `resolved_from` in the result |
| `-32602`, `error_reason: "invalid_args"`, `checked_before_payment: true` | An unknown, missing or malformed parameter, refused before any challenge. You were not charged, and the message names what it expected. `search_lps` with `limit` above 25 lands here |
| `-32001`, `error_reason: "missing_key"` | The tool needs a key. The message opens with how a person gets one (https://fundmomentum.vc/pricing), then the agent route (`POST` to the register URL), then which tier the tool needs. The same two-part opening is used by `anon_quota_exceeded`, `anon_ip_ceiling_exceeded`, `insufficient_tier` and the `-32042` message; fields in `error.data` are unchanged |
| `-32001`, `error_reason: "pull_quota_exceeded"` | `get_fund` or `search_lps` on a Free/Starter key that has used its monthly profile-pull allowance (they share one pool). No challenge, nothing charged; `error.data` carries `pulls_used`, `pulls_limit`, `resets_at` and an upgrade link |
| `-32001` (HTTP 429), `error_reason: "fair_use_daily_lp_rows"` | `search_lps` on a Pro key that has been served 500 LP rows today (UTC). Pro pulls stay unlimited per month; this only stops runaway loops. `error.data` carries `lp_rows_today`, `daily_limit`, `next_step` and `resets_at` (also `Retry-After`). Nothing charged |

## Closes: last_close, closed_since, close_updated

Fund Momentum publishes fund closes as primary research, and each covered close
is written back into the fund record:

- `get_fund` returns `last_close`: `type` (launch, first, interim, final,
  closed_unspecified, target_only), `date`, `amount_m` and `currency` exactly
  as the fund stated them (never converted), `size_display`, `target_m`,
  `source_url` (our post) and `primary_url` (the press release). `null` means
  we have not covered a close of that fund, not that it has not closed.
  `date` is the day Fund Momentum reported the close (`date_basis: "reported"`,
  see `date_basis_note`), not necessarily the day the fund signed it.
- `search_funds` with `closed_since: "2026-09-01"` returns funds whose latest
  covered close is on or after that date, newest first, with `lastCloseDate`
  and `lastCloseType` on every result. `launch` and `target_only` entries are
  skipped.
- Write-back rules: a post applies only to the fund whose vintage it names (no
  numeral means Fund I); posts covering several funds never change a fund's
  size; closes found only on the open web are not written into fund records.
- `get_changes` marks those rows `change_type: "close_updated"` with a
  `last_close` summary.

## Webhooks

Push instead of poll. Available on every key plan, Free included. Agent credit keys have no webhooks: poll `get_changes`.

- Free and Starter keys: 1 webhook subscription, filters allowed, up to 50 delivered events per month (UTC calendar month, counted per key).
- Pro and Enterprise: up to 5 webhook subscriptions per key, unlimited events.
- Over the monthly allowance, deliveries pause (the subscription is kept) until the next UTC month starts or you upgrade at https://fundmomentum.vc/pricing; GET /_api/webhooks/list reports delivery_paused_until.

Subscribe an HTTPS endpoint and every newly published
close arrives as a signed `close_updated` event:

```bash
curl -s -X POST https://fundmomentum.vc/_api/webhooks/subscribe \
  -H 'X-API-Key: YOUR_KEY' -H 'Content-Type: application/json' \
  -d '{"url":"https://example.com/fm-hook","filters":{"stage":"seed","country":"Germany"}}'
```

The response carries the signing `secret` once. Every delivery has
`FM-Signature: t=<unix>,v1=<hex>` where `v1` is HMAC-SHA256 of `"<t>.<raw body>"`
with that secret; reject anything older than 5 minutes. Also sent: `FM-Event`,
`FM-Event-Id` (stable per change, use it to de-duplicate) and `FM-Delivery`.
Answer 2xx within 10 seconds. Failures are retried after 1 min, 5 min, 30 min,
2 h, 6 h, 12 h and 24 h; 40 failed attempts in a row switch the subscription
off, and a successful `test` switches it back on. Redirects are not followed.

| Endpoint | Body | Does |
|---|---|---|
| `POST /_api/webhooks/subscribe` | `url`, optional `filters.stage/country/industry`, `description` | Create; returns the stable `subscription_id`. Over the plan's subscription limit: 403 with an upgrade `next_step` |
| `GET /_api/webhooks/list` | | Plan allowance, `delivery_paused_until`, subscriptions (`subscription_id`, `url`, `filters`, `created_at`, `events_this_month`), last 20 deliveries |
| `POST /_api/webhooks/test` | `subscription_id` | Send a signed `ping` now |
| `POST /_api/webhooks/delete` | `subscription_id` or `url` | Remove |

Event body:

```json
{
  "id": "evt_close_123_1791830000000",
  "type": "close_updated",
  "api_version": "2026-10-12",
  "created_at": "2026-10-12T09:00:00.000Z",
  "data": {
    "fund": { "slug": "...", "name": "...", "country": "Germany", "fundingStage": "seed",
              "fundSize": "€33.3M", "raisingStatus": "closed", "industries": ["saas"],
              "url": "https://fundmomentum.vc/funds/..." },
    "last_close": { "type": "final", "date": "2026-09-01", "amount_m": 33.3, "currency": "EUR",
                    "size_display": "€33.3M", "target_m": null,
                    "source_url": "https://fundmomentum.vc/blog/...", "primary_url": "https://..." },
    "previous_close": { "type": "first", "date": "2026-03-01" }
  }
}
```

## REST equivalents (Make, Zapier, n8n, HubSpot)

Every tool is also a plain REST endpoint. They call the same handler as
`POST /_api/mcp`, so validation, auth (`X-API-Key` or `Authorization: Bearer`;
keyless where the tool is keyless), quotas, pricing and the MPP 402 are
identical. There is no JSON-RPC envelope: the response is the tool result
with `_meta` (usage, credits) as a top-level field. Floot endpoints take no
path parameters, so a fund slug is a query parameter.

| REST | Mirrors | Parameters |
|---|---|---|
| `GET /_api/v1/funds` | `search_funds` | `stage`, `country`, `industry`, `closed_since`, `limit`, `offset` |
| `GET /_api/v1/funds/detail` | `get_fund` | `slug` |
| `GET /_api/v1/funds/signals` | `get_fund_signals` | `slug` |
| `GET /_api/v1/funds/gps` | `get_gp_profile` | `slug`, optional `gp_name` |
| `POST /_api/v1/match` | `match_startup` | JSON body: `description`, `country`, `stage` |
| `GET /_api/v1/lps/coverage` | `check_lp_coverage` | `country`, `lp_type` |
| `GET /_api/v1/lps` | `search_lps` | `country`, `lp_type`, `backs_emerging_managers`, `limit` (max 25) |
| `GET /_api/v1/changes` | `get_changes` | `since`, `limit`; `ETag` / `If-None-Match` (304 when unchanged) |
| `GET /_api/v1/usage` | `get_usage` | none |
| `GET /_api/v1/enums` | n/a | Accepted values for `stage`, `industry`, `lp_type` (for dropdowns) |

`/_api/agent/me` and the older `/_api/changes` keep working.

- REST v1 contract: fields are only ever added, never renamed or removed. Breaking changes ship as /v2.
- REST v1 search_funds: default limit 10; ceiling per call 50 for keyless, Free, Starter and agent-credit callers and 250 for Pro and above. Keyless and free-key callers are held to 10 by the plan, exactly as over MCP, and agent credits keep the existing cap of 150 funds per key per UTC day. A limit above the ceiling is rejected with 400, never clamped. MCP limits are unchanged.
- A 304 and an empty poll are both free. A conditional request that answers 304 Not Modified (If-None-Match matches the current ETag; over MCP, get_changes answers unchanged: true) is not counted against keyless, Free, Starter or Pro call quotas and is not billed to agent credits. It still counts against the per-minute pace limit. A request whose ETag no longer matches returns the changes and is counted as one normal call.
- Polls that return no changes are free: they do not count against quotas or credits (the per-minute pace limit still applies). Keyless get_changes callers are held to 30 calls per minute and Free keys to their existing per-minute limit. There is no daily cap on empty polls. A response reports _meta.billed (false for an empty poll or a 304, true otherwise).
- Automation clients may send X-Client (for example X-Client: make). It is informational, appears in usage statistics only and changes no behaviour.
- Every REST v1 error is { error: { code, message, error_reason, next_step } } with a real HTTP status. 402, 403 and 429 always carry a human-readable next_step, and 429 also carries resets_at (and a Retry-After header) where a reset time exists.
- Changelog: https://fundmomentum.vc/_api/api-changelog

Errors are an HTTP status plus
`{"error":{"code","message","error_reason","next_step"}}`:
400 `invalid_args`, 404 `not_found`, 401 `missing_key` / `invalid_key`,
403 `insufficient_tier`, 429 `pull_quota_exceeded` / `anon_quota_exceeded` with `resets_at`. A Pro
tool called without a key or credits answers a real `402` with a
`WWW-Authenticate: Payment ...` challenge. Billing is the MCP billing:
`search_funds` is 1 credit per fund returned (an empty result is free), default `limit` 10.
Optional `max_credits` (1 to 25) caps the page and never raises `limit`; every response
carries a `billing` object (`credits_charged`, `balance_after`, `next_page_cost`,
`next_offset`, `payment_mode`, `credits_unreturned`). Too few credits returns the rows you
can afford; none returns `402` as `application/problem+json` with `required`,
`available`, `shortfall` and `top_up`. Machine-readable
spec: https://fundmomentum.vc/openapi.json

## Provenance

Every `get_fund` response carries a `provenance` block: where each field came
from, when it was last verified, and a confidence level. Fields we have not
verified say so rather than guessing. `provenance.field_sources` names the
source of fund size, stage (with `fundingStage_confidence`) and GP names, ranked
claim (fund-verified) > admin > blog (our close coverage) > website > llm.
GP names are kept only when they appear in a fetched source with a first and
last name.

---

Questions, or a failed run worth telling us about: michael@fundmomentum.vc
