# Foresea Forecasting (remote · foresea.ink)

Forecast future events and scan prediction-market edges.

- Trust score: 67/100 (medium)
- Change this week: +5
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-08-04

## Components

- remote · `foresea.ink`: 67/100 (this document), [markdown](https://verifymcp.io/servers/ink-foresea-forecasting/foresea.md), [page](https://verifymcp.io/servers/ink-foresea-forecasting/foresea)

## Channel facts

- Endpoint: `https://foresea.ink/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 2026-08-04.

- **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 5 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**: 76/100
  - 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).
  - AI-judged instruction clarity (good).
  - Context-footprint check failed: tool/resource definitions use about 1095 tokens (~156/item across 7 items; 5 tools + 2 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**: 71/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 0% of tool parameters carry a description.
  - Structured output schemas are declared (100% of tools); any adoption earns full credit.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### Claude

```bash
claude mcp add --transport http ink-foresea-forecasting https://foresea.ink/mcp/
```

### Codex

```toml
[mcp_servers.ink-foresea-forecasting]
url = "https://foresea.ink/mcp/"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "ink-foresea-forecasting": {
      "type": "remote",
      "url": "https://foresea.ink/mcp/",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add ink-foresea-forecasting --url https://foresea.ink/mcp/ --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  ink-foresea-forecasting:
    url: "https://foresea.ink/mcp/"
```

### Other

```json
{
  "mcpServers": {
    "ink-foresea-forecasting": {
      "type": "http",
      "url": "https://foresea.ink/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.

### 2026-08-03 (score 67, +1)

No change was recorded against any check on this day. Stability & Change Management went from 23 to 27. That category is still filling its 30-day observation window: 7 days of observed history at the previous scan, 8 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-08-01 (score 66, +41)

- [security improvement] Transport: fail → pass
- [security] Authorization: Authorisation not fully verified: no authorisation is required to call this server, and 5 tool(s) never declared a destructiveHint. The MCP spec treats an absent hint as destructive by default, so we cannot call this surface safe.
- [functional improvement] Endpoint reachability: not serving MCP → reachable
- [functional improvement] MCP protocol: unverified → pass
- [functional improvement] Schema quality: unverified → 100
- [functional improvement] Stability: unverified → 0.20
- [functional improvement] Tool coverage: unverified → 100
- [functional] Server version: 1.28.1 → 1.29.0

### 2026-07-31 (score 25, +9)

- [functional] We updated how we score, so this day's move reflects our rubric, not a change to the server

### 2026-07-30 (score 16, 0)

- [functional] We updated how we score, so this day's move reflects our rubric, not a change to the server

### 2026-07-29 (score 16, −46)

- [security regression] Endpoint reachability: reachable → not serving MCP
- [security regression] Stability: 0.07 → unverified
- [security regression] Transport: pass → fail
- [security] Authorization: Authorisation not fully verified: no authorisation is required to connect, but we couldn't read the tool list to see what that exposes.
- [functional regression] Tool coverage: 100 → unverified
- [functional regression] Capabilities: pass → unverified
- [functional regression] Schema quality: 100 → unverified

### 2026-07-28 (score 62, +1)

No change was recorded against any check on this day. Stability & Change Management went from 3 to 7. That category is still filling its 30-day observation window: 1 days of observed history at the previous scan, 2 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-07-27 (score 61, +1)

- [functional] We updated how we score, so this day's move reflects our rubric, not a change to the server

### 2026-07-26 (score 60)

First indexed and scored.

## MCP tools (5)

### `foresea_forecast` (~313 tokens)

Call this whenever the user asks about probability, likelihood, or whether
        something will happen. Good triggers: "Will X happen?", "What are the chances
        of Y?", "How likely is Z?", "What's the probability that…", "Do you think X
        will…", "Should I bet on…". Returns a calibrated YES/NO probability (or
        numeric/date range) with written rationale and supporting news evidence.
        If you also have a market price (market_probability) or URL (market_url), pass
        it to get the model-vs-market edge — how mispriced the market is.
        Example: question="Will the Fed cut rates by March 2026?", market_probability=0.4
        → {predicted_answer:"No", confidence:0.62, rationale, evidence_sources,
        market_analysis:{model_probability:0.54, edge:+0.14, stance:"model_above_market"}}
        Handles: binary YES/NO, multiple-choice, numeric ranges, and date questions.

Input parameters:

- `attach_evidence` (boolean)
- `categories`
- `description` (string)
- `evidence_top_k` (integer)
- `market_outcome`
- `market_platform`
- `market_probability`
- `market_url`
- `options`
- `question` (string, required)
- `question_type`
- `resolution_criteria` (string)
- `variant` (string)

Output parameters:

- `result` (object)

### `foresea_analyze_market` (~260 tokens)

Call this when the user mentions a specific prediction market by URL, slug,
        or ticker — or asks whether a particular market is over/underpriced. Good
        triggers: "Is this Polymarket fair?", "What's the edge on kalshi:XXXXX?",
        "Should I buy/sell this market?", user pastes a Polymarket or Kalshi URL.
        Fetches the live price, gathers evidence, forecasts, computes model-vs-market
        edge, and returns a recommendation. Use foresea_forecast instead when there is
        no specific live market — just a general probability question.
        Example: platform="polymarket", slug="fed-rate-cut-march-2026"
        → {model_probability, market_probability, edge, stance, recommendation, thesis}.

Input parameters:

- `builtin_skills` (boolean)
- `evidence_top_k` (integer)
- `ground_in_record` (boolean)
- `market_id`
- `market_probability`
- `max_tool_steps` (integer)
- `platform`
- `question`
- `skills`
- `slug`
- `ticker`
- `tool_loop` (boolean)
- `variant` (string)

Output parameters:

- `result` (object)

### `foresea_scan_markets` (~186 tokens)

Call this when the user wants to find mispriced or interesting markets, not
        evaluate a specific one. Good triggers: "What should I bet on?", "Find me
        trading opportunities", "Which markets are mispriced right now?", "What's
        Foresea's best edge today?", "Scan Polymarket for opportunities".
        Returns markets ranked by model-vs-market disagreement, each with model
        probability, market price, and edge. For a specific market, use
        foresea_analyze_market instead.
        Example: platform="kalshi", min_edge=0.1
        → [{question, market_probability, model_probability, edge, market_url}].

Input parameters:

- `evidence_top_k` (integer)
- `limit` (integer)
- `min_edge` (number)
- `platform` (string)
- `query`

Output parameters:

- `result` (object)

### `foresea_track_record` (~96 tokens)

Call this when the user asks how reliable or accurate Foresea is, or wants
        to know whether to trust a forecast. Good triggers: "How good is Foresea?",
        "What's the track record?", "Has it been right before?", "Is it calibrated?",
        "What's the Brier score?". Returns accuracy, Brier score, calibration (ECE),
        and skill-vs-market broken down by time horizon.

Output parameters:

- `result` (object)

### `foresea_edge_board` (~118 tokens)

Call this when the user wants the current top trading opportunities with
        explicit trade directions and historical backing. Good triggers: "What are
        the best bets right now?", "Show me the edge board", "Which model is winning
        the paper-trading competition?", "What's the strongest edge today?",
        "Are these edges statistically significant?". Returns open markets ranked by
        model-vs-market disagreement, each with Buy YES/NO direction, implied odds,
        whether the edge is historically significant, and a multi-model comparison.

Output parameters:

- `result` (object)

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/ink-foresea-forecasting/foresea#diagnostics

## Score history

- 2026-08-04: 67
- 2026-08-03: 67
- 2026-08-02: 66
- 2026-08-01: 66
- 2026-07-31: 25
- 2026-07-30: 16
- 2026-07-29: 16
- 2026-07-28: 62
- 2026-07-27: 61
- 2026-07-26: 60

## Links

- Remote endpoint: https://foresea.ink/mcp/
- Repository: https://github.com/pareelamre/analyzing-llm-rationale
- Website: https://foresea.ink/
- Changelog RSS feed: https://verifymcp.io/servers/ink-foresea-forecasting/foresea/changelog.xml
- Changelog JSON feed: https://verifymcp.io/servers/ink-foresea-forecasting/foresea/changelog.json
- HTML version of this page: https://verifymcp.io/servers/ink-foresea-forecasting/foresea