# IRIS: startup intelligence (remote · iris-media.xyz)

Sourced startup intelligence: SEC Form D rounds, launches, tech wire, market and startup jobs.

- Trust score: 67/100 (medium)
- Change this week: +4
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-21

## Components

- remote · `iris-media.xyz`: 67/100 (this document), [markdown](https://verifymcp.io/servers/xyz-iris-media-iris/api-mcp.md), [page](https://verifymcp.io/servers/xyz-iris-media-iris/api-mcp)

## Channel facts

- Endpoint: `https://iris-media.xyz/api/mcp`
- Transports: `streamable-http`
- Auth: `none`
- Version: `1.0.0`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-11-21.

- **Endpoint Security**: 63/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - Authorisation not fully verified: no authorisation is required to call this server, and 17 tool(s) never declared a destructiveHint. The MCP spec treats an absent hint as destructive by default, so we cannot call this surface safe.
  - HTTPS is enforced; there's no plaintext access path.
  - The HSTS (Strict-Transport-Security) header is present.
  - DNSSEC check failed: this domain isn't protected by DNSSEC.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 67/100
  - AI-judged instruction clarity (good).
  - Context-footprint check failed: tool/resource definitions use about 1926 tokens (~113/item across 17 items; 17 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 30/100
  - Stability observed for 9 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 83/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 48% of tool parameters carry a description.
- **Tool Safety**: 75/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - 0 of 1 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation; "send_feedback" implies "send" and declares no destructiveHint at all, which the MCP spec reads as destructive by default.
  - An AI judge read all 18 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### How do I install the IRIS: startup intelligence MCP server?

IRIS: startup intelligence is a hosted endpoint at https://iris-media.xyz/api/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add --transport http xyz-iris-media-iris 'https://iris-media.xyz/api/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "xyz-iris-media-iris": {
      "url": "https://iris-media.xyz/api/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "xyz-iris-media-iris": {
      "type": "http",
      "url": "https://iris-media.xyz/api/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.xyz-iris-media-iris]
url = "https://iris-media.xyz/api/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "xyz-iris-media-iris": {
      "type": "remote",
      "url": "https://iris-media.xyz/api/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add xyz-iris-media-iris --url 'https://iris-media.xyz/api/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  xyz-iris-media-iris:
    url: "https://iris-media.xyz/api/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "xyz-iris-media-iris": {
      "Transport": "http",
      "Url": "https://iris-media.xyz/api/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add xyz-iris-media-iris -t streamable-http -u 'https://iris-media.xyz/api/mcp'
```

### Other

```json
{
  "mcpServers": {
    "xyz-iris-media-iris": {
      "type": "http",
      "url": "https://iris-media.xyz/api/mcp"
    }
  }
}
```

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-11-21 (score 67, +1)

No change was recorded against any check on this day. Stability & Change Management went from 27 to 30. That category is still filling its 30-day observation window: 8 days of observed history at the previous scan, 9 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-19 (score 66, +1)

No change was recorded against any check on this day. Stability & Change Management went from 20 to 23. That category is still filling its 30-day observation window: 6 days of observed history at the previous scan, 7 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-17 (score 65, +1)

No change was recorded against any check on this day. Stability & Change Management went from 13 to 17. That category is still filling its 30-day observation window: 4 days of observed history at the previous scan, 5 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-16 (score 64, 0)

- [functional] New tool “get_events”

### 2026-11-15 (score 64, +1)

No change was recorded against any check on this day. Stability & Change Management went from 7 to 10. That category is still filling its 30-day observation window: 2 days of observed history at the previous scan, 3 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-14 (score 63, 0)

- [functional regression] Schema quality: pass → fail
- [functional] New tool “ask”
- [functional] New tool “get_alert_matches”
- [functional] New tool “get_company_filings”
- [functional] New tool “get_rising”
- [functional] New tool “get_startups”

### 2026-11-13 (score 63, +1)

- [functional improvement] Stability: unverified → 0.03

### 2026-11-12 (score 62)

First indexed and scored.

## MCP tools (17)

### `thesis_report` (~157 tokens)

Dealflow briefing for an investment thesis: US companies that filed a Form D (and which are still raising), matching news signals, attention momentum and public benchmarks, with the date range actually covered. Every item has a source URL. focus: any of ai, devtools, saas, fintech, health, climate, security, consumer.

Input parameters:

- `cities` (array): e.g. San Francisco, New York, Boston. Empty = all US
- `days` (number): 7-30, default 14
- `focus` (array)
- `match` (string): strong (default): only filings with evidence for the focus. possible: also technology filings with no evidence (Form D has no AI or devtools category).

### `get_rounds` (~103 tokens)

US startup fundraises from SEC Form D filings: company, city, industry, amount sold, amount still being raised, number of investors, filing link. One row per company (latest filing, with a `filings` count). Returns the date range actually covered. Pass a 2-letter state for full 14-day coverage of that state.

Input parameters:

- `city` (string)
- `state` (string): 2-letter US state, e.g. CA

### `get_launches` (~77 tokens)

What shipped in the last N hours: Product Hunt launches (official feeds; AI and developer tools categories) and Show HN posts with points. Each item links to its launch page; Show HN items include the product URL.

Input parameters:

- `category` (string)
- `hours` (number): 6-72, default 48

### `get_wire` (~95 tokens)

Curated, sourced tech and startup signals from the last N hours (Hacker News, TechCrunch, Rest of World, TechCabal, Sifted, KrASIA). Headlines are the sources' own; each links to the original.

Input parameters:

- `hours` (number): 1-72, default 24
- `raw` (boolean): true returns everything collected, uncurated. Default: curated for startup intelligence.

### `get_voices` (~116 tokens)

Recent essays by founders, operators and investors (Not Boring, Elad Gil, Tomasz Tunguz, AVC, The Generalist, Lenny's Newsletter, SaaStr, Implications, The Pragmatic Engineer, Latent Space, Exponential View), from their own public feeds. Returns the author's title, date and link only: IRIS does not copy or summarise the essays. Opinion, not verified fact.

Input parameters:

- `days` (number): 1-30, default 14
- `role` (string)

### `get_rising` (~113 tokens)

Emerging startups, daily: rounds announced in the last 72 hours in the startup press of the US, Europe, Latin America (Spanish and Portuguese too), India, China, Southeast Asia, Canada and Africa (company, amount, stage, notable backers, source link), plus Y Combinator launches from the last 14 days. Read from headlines; items with an unreadable company name are skipped, never guessed.

Input parameters:

- `region` (string): Middle East includes Israel and Türkiye; Asia includes Japan and Korea

### `ask` (~105 tokens)

Ask IRIS a plain question in English, Spanish or Portuguese ("who founded Nubank?", "which startups raised in Latin America this week?", "startups líderes en México", "what did Naval say about wealth?"). Routes to the right IRIS dataset and answers with original source links (no generated prose), plus cite_as: how to credit IRIS when you repeat it.

Input parameters:

- `lang` (string)
- `q` (string, required): The question, up to 300 characters

### `get_events` (~154 tokens)

Upcoming startup, AI and founder events in 47 cities worldwide, read from Luma's public city pages (refreshed hourly) and filtered for startup topics. Each event has its start time in the local time zone, city, host and the Luma link where the organizer takes registrations. Filter by city slug (e.g. bogota, sf, london), country name (e.g. Colombia) or region.

Input parameters:

- `city` (string): Luma city slug, e.g. bogota, mexico-city, bangalore
- `country` (string): Country name in English, e.g. Colombia
- `days` (number): 1-90, default 30
- `region` (string)

### `get_alert_matches` (~124 tokens)

Agent alerts: what matched a filter in the last 24 hours. Rounds announced in the startup press (and, with follow=all, Front Page stories) for a market and/or keyword (country, sector or company; accent-insensitive whole-word prefix, so "Colombia" matches "Colombian"). Call it on a schedule to watch a market; the same filters power IRIS email alerts (iris-media.xyz/alerts).

Input parameters:

- `follow` (string)
- `keyword` (string): Up to 40 characters
- `market` (string)

### `get_company_filings` (~136 tokens)

Form D history for one US company: every Form D (new offering) and D/A (amendment) it filed with the SEC, newest first, with amount sold, amount offered, number of investors and first sale date per filing. Answers "has this startup raised before, and how often?". Pass the CIK or the sec_url from get_rounds. Amounts are per filing and never summed (an amendment updates an earlier offering). Form D does not report valuation.

Input parameters:

- `cik` (string): SEC Central Index Key, digits only
- `sec_url` (string): A filing URL from get_rounds

### `get_startups` (~113 tokens)

The IRIS Startup Atlas: the leading private startups of North America, Latin America, Europe, Africa, the Middle East, China, India and Southeast Asia. Each profile has founders, a one-line description (English and Spanish), dated facts with source URLs, and a founder quote copied verbatim from a checked source (original language, with an English translation). Verified by hand; check verified_on.

Input parameters:

- `region` (string)
- `sector` (string): e.g. fintech, ai, commerce, robotics

### `get_market` (~68 tokens)

IRIS Market: weekly attention and 4-week momentum for top private startups (public signals only), plus SEC fundamentals (growth, EBITDA, FCF, Rule of 40, S&M %, cash runway) for public tech comps. Optional tickers filter.

Input parameters:

- `tickers` (array)

### `plan_for_project` (~64 tokens)

Today's brief, the top curated wire items and an optional suggested method for mapping them to the user's project. The method is a suggestion, not an instruction. Only sourced items count as facts.

Input parameters:

- `project` (string): Short description of the user's project, stack and goals.

### `get_jobs` (~160 tokens)

Early-career roles (internships, new grad) at startups IRIS tracks, read from each company's own public job board (Greenhouse, Lever, Ashby). Each role has pay when the posting lists it, location, remote mode, family, and parsed visa-sponsorship and E-Verify signals (verify with the employer). Always send the user to `url`, the employer's own page.

Input parameters:

- `city` (string)
- `early_career_only` (boolean)
- `family` (string)
- `limit` (number): 1-50, default 20
- `paid_only` (boolean)
- `query` (string)
- `remote` (string)
- `sponsors_visas` (boolean)

### `send_feedback` (~218 tokens)

Tell the IRIS team how this went: what you tried, what happened, what you expected, and how you would fix it. Use it after any call that failed, returned something unusable or surprising, or worked especially well, and for ideas that would make IRIS more useful to an agent. Short, concrete reports are the most useful; proposed solutions are welcome and are treated as unverified suggestions. A person reads every report and nothing is executed automatically. Do not include secrets or personal data about your user.

Input parameters:

- `agent` (object)
- `contact_email` (string): Only if you want a reply
- `expected` (string)
- `kind` (string)
- `proposed_solution` (string)
- `severity` (string)
- `summary` (string, required): One sentence, 12-300 characters
- `tool` (string): The IRIS tool this is about, e.g. get_rounds
- `what_happened` (string)
- `what_i_tried` (string)

### `get_daily_brief` (~40 tokens)

Editor-approved IRIS briefs (signal, meaning, forecast with confidence). May be empty on days without approved briefs; use get_wire for sourced signals.

### `get_upcoming` (~28 tokens)

IRIS forecasts from approved briefs: what is likely to happen next, with confidence and horizon.

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/xyz-iris-media-iris/api-mcp#diagnostics

## Score history

- 2026-11-21: 67
- 2026-11-20: 66
- 2026-11-19: 66
- 2026-11-18: 65
- 2026-11-17: 65
- 2026-11-16: 64
- 2026-11-15: 64
- 2026-11-14: 63
- 2026-11-13: 63
- 2026-11-12: 62

## Common questions

### What is the IRIS: startup intelligence MCP server?

IRIS: startup intelligence is an MCP server listed in the public MCP registry as xyz.iris-media/iris. Sourced startup intelligence: SEC Form D rounds, launches, tech wire, market and startup jobs. This page covers its hosted endpoint (https://iris-media.xyz/api/mcp).

### Is the IRIS: startup intelligence MCP server safe to use?

IRIS: startup intelligence scores 67 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the IRIS: startup intelligence MCP server expose?

IRIS: startup intelligence exposes 17 tools: thesis_report, get_rounds, get_launches, get_wire, get_voices, and 12 more. Their descriptions and schemas cost roughly 1,871 tokens of context every time the server is loaded.

### Does the IRIS: startup intelligence MCP server require authentication?

No. We connected to IRIS: startup intelligence without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

### Is the IRIS: startup intelligence MCP server still maintained?

IRIS: startup intelligence is still listed as active in the MCP registry. We last reached this channel on 21 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

### What did VerifyMCP test on the IRIS: startup intelligence MCP server?

VerifyMCP ran 12 checks on IRIS: startup intelligence, covering TLS certificate, HTTPS, HSTS header, DNSSEC and 8 more. One item could not be completed, a gap on our side rather than a finding about IRIS: startup intelligence: Authorization. Supply Chain Security and Provenance & Transparency do not apply to a hosted endpoint, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.

## Links

- Remote endpoint: https://iris-media.xyz/api/mcp
- Website: https://iris-media.xyz/docs
- Changelog RSS feed: https://verifymcp.io/servers/xyz-iris-media-iris/api-mcp.xml
- Changelog JSON feed: https://verifymcp.io/servers/xyz-iris-media-iris/api-mcp/changelog.json
- Trust report (JSON): https://verifymcp.io/servers/xyz-iris-media-iris/api-mcp.json
- HTML version of this page: https://verifymcp.io/servers/xyz-iris-media-iris/api-mcp
