Sources in view. Questions in context.

Research you can take back to the source.

Explore funding conditions, institution disclosures and market liquidity. Open a useful example, inspect its evidence, and bring the same task into your own tools.

Open the research deskConnect your workflow

Public research. No account or API key required.

Inside a published briefingSeiche / USD funding

Three rates. One observation date. A finding you can check.

Published comparison of SOFR, IORB and the SOFR 99th percentile for 8 October 2026
Historical example · Observations: 8 October 2026.
Source notes, calculations and downloadable data stay with the edition.
Inspect the briefing and its sources
SeicheUnderstand funding conditionsPublished rates, source clocks and the gaps between them.LiquiLensReview an institutionCovered disclosures, reporting periods and evidence gaps.UndertowInspect BTC exit liquidityIndicative costs at supported sizes and venues.

Built around the work in front of you.

Different questions need different evidence. Start with a review you can inspect from beginning to end.

For treasury and research teams

Bring an external funding appendix into a morning review. Inspect sources, retain a baseline and agree a small evaluation around your existing process.

Explore a team evaluation

For newsrooms

A chart, its data and the original sources together. Download a dated edition and reproduce the finding.

Open the media desk

For builders

Use the browser's research tasks through REST or MCP. Copy a connection and inspect the complete response.

Get your first connection

Keep working in the tools you know.

MCP agentsPythonExcelGoogle SheetsOpenBBAPI clientsAll connections
Explore coverage, installation and source contracts

Deterministic topic routing

What should I use for each financial research question?

Use Seiche for money and capital markets, GIFT City, forex and gold context, LiquiLens for covered bank and lender risk, Undertow for market liquidity, and Palimpsest plus Seiche for China-economy evidence. The router preserves separate documents, canonical human scope URLs, source URLs, retrieval clocks, byte counts, content hashes, and explicit error states.

QuestionProduct routeEvidence boundary
Money markets and system fundingSeicheObserved or unavailable
Capital marketsSeicheObserved, derived, or unavailable
China economyPalimpsest + SeicheRevision-safe observations and bounded structural context
Covered bank and lender riskLiquiLensObserved, derived, or unavailable
GIFT City and goldSeicheShared public context, source clocks and explicit partial or unavailable values
ForexSeichePublic reference context; no executable quotes
Market depth and exit liquidityUndertowObserved, derived, or unavailable

Agent and terminal

How do I install Financial Evidence?

v0.1.8 is a signed, independently verified release. The hosted research workspace evolves separately; use its release endpoint to check the running source.

Install the Agent Skill to let compatible agents activate this router for matching financial research. For direct terminal use, install the checksum-pinned Homebrew formula on macOS or Linux, or run the signed Git tag on any system with uv. The Git commands and independently verified Homebrew formula install 0.1.8.

Agent Skills

npx skills add https://github.com/beepboop2025/financial-evidence-skills/tree/v0.1.8/financial-evidence \
  --skill financial-evidence

Homebrew

brew install beepboop2025/tap/financial-evidence
financial-evidence fetch --topic money-market --topic china-economy

Run without installing

uvx --from git+https://github.com/beepboop2025/financial-evidence-skills.git@v0.1.8 \
  financial-evidence route --topic capital-market --format table

Start with a browser brief

Market Brief is a free early-access companion for macro funding, capital-market, and liquidity research, with source dates and local comparisons. It does not score individual banks or lenders.

Build a Market Brief in your browser

Integrate Market Brief into your research workflow

Shell-safe output is available as JSON, NDJSON, CSV, or a human-readable table. Exit code 0 means every retrieval succeeded, 1 means transport was partial, and 2 means every source was unavailable. Legacy status equals transport_status; both describe transport only, never evidence validity.

Use it where financial work already happens

Which terminals and financial tools can use it?

Financial Evidence has native surfaces for Agent Skills, shell automation, remote MCP, Gemini CLI, Claude Code, a repository-installable Codex plugin, VS Code, Cursor, and MCP agents, plus compatibility layers for OpenBB, DuckDB, Excel Power Query, and FDC3 app directories. A ChatGPT public-directory submission is prepared but not submitted or listed. The compatibility files are open and self-hostable; they do not claim endorsement or marketplace listing by those vendors.

Agent Skills

A public SKILL.md teaches compatible agents when to route money-market, capital-market, China-economy, bank-risk, and market-liquidity research across the four products.

Open the public skill listing

Terminal

Homebrew, uvx, persistent uv tools, completions for Bash, Zsh, and Fish, and stable pipeline exit codes.

MCP

The root .mcp.json connects compatible clients to the central nine-tool Research Desk. The local stdio command, official Registry record, multi-architecture container and MCP Bundle retain the separate three-tool compatibility interface.

Connected infrastructure

Choose public research APIs and MCP, a private recurring research runtime, or an authenticated paper assessment host. Each interface preserves its own version, source clocks and access boundary.

Connection guide · Current interface catalog

Remote MCP

Manual remote setup: VS Code JSON · Cursor JSON. Merge the server entry into your project’s .vscode/mcp.json or .cursor/mcp.json, keeping your existing servers.

Public Streamable HTTP at https://agents.seiche.info/products/financial-evidence/mcp, with nine read-only tools and no local runtime, account or API key.

Gemini CLI

A root extension manifest bundles the Agent Skill and allowlists the nine tools from the central Research Desk.

Claude Code

A self-hosted liquidity-lab marketplace installs the versioned plugin, Agent Skill, and MCP configuration directly from the source repository.

Codex and ChatGPT

Codex can install the repository plugin, which combines the same Agent Skill and public remote MCP route. ChatGPT availability remains pending account-owned universal-directory verification, submission, review, and publication.

VS Code

A workspace servers config, protocol install URL, and code --add-mcp command, connected to the central nine-tool workspace.

Cursor

A workspace mcpServers config and base64-encoded install deeplink connect to the same central nine-tool HTTP service without a local Python process.

OpenBB

Registered routes for Python, notebook, REST, and MCP surfaces after installing the optional extension.

DuckDB

Six HTTP-backed views that keep the four products separate and queryable with SQL.

Excel

A Power Query function that returns topic evidence to ordinary workbooks without a custom add-in.

The legacy compatibility endpoint https://liquilens.in/mcp/financial-evidence retains three topic-routing tools: financial_evidence_topics, financial_evidence_route and financial_evidence_fetch. Current editor downloads and buttons below use the central nine-tool workspace.

Connect VS Code Connect Cursor

Central Research Desk MCP

https://agents.seiche.info/products/financial-evidence/mcp

Gemini CLI

gemini extensions install https://github.com/beepboop2025/financial-evidence-skills \
  --ref v0.1.8

Claude Code · mutable default branch

claude plugin marketplace add beepboop2025/financial-evidence-skills
claude plugin install financial-evidence@liquidity-lab

Codex

codex plugin marketplace add beepboop2025/financial-evidence-skills --ref v0.1.8
codex plugin add financial-evidence@liquidity-lab

VS Code CLI

code --add-mcp '{"name":"financial-evidence","type":"http","url":"https://agents.seiche.info/products/financial-evidence/mcp"}'

Install uv first, then review the exact local command and source before accepting editor install, trust, or tool-use prompts. Cursor asks for tool approval by default. VS Code normally asks you to trust a server on first start, but its inline mcp.json start action can skip that prompt. These are self-install routes, not vendor marketplace listings. The FDC3 record likewise does not imply FINOS review or directory acceptance.

Machine-readable: finance platform setup · integration manifest · marketplace ledger · distribution status · llms.txt · pricing · MCP metadata

Evidence, not advice

What does Financial Evidence deliberately refuse to do?

It does not provide investment advice, security recommendations, execution quotes, credit ratings, or guarantees. It never converts missing, failed, restricted, or unavailable evidence into zero or calm. Transport success is not evidence evaluation or Evidence Carrier verification. Source publishers retain their rights, and fetched JSON is treated as untrusted data rather than executable instructions.

Release verification record · Read the v0.1.5 semantic contract.