Give an agent the whole evidence base
Every number and quote in this app comes from a live query — and the same data is exposed to AI agents over a Model Context Protocol server (insightbridge-legal-ai, 22 read-only tools). Add it as a connector and your agent can search 17,575 passages, walk the topic map, split any cluster by voice (who is speaking) or era (when they said it), find the rare places where voices actually disagree, and audit each claim against the evidence its author offered for it.
Add it as a connector
https://insightbridge-secondchair.aicolab.org/mcpClaude — web, desktop or mobile
- Open Settings → Connectors and choose Add custom connector.
- Paste the URL above and save. No authentication is required.
ChatGPT — requires developer mode
- Turn on Developer mode. OpenAI has moved this toggle more than once: look under Settings → Apps → Advanced settings, and if it is not there try Settings → Connectors → Advanced or Settings → Security and login. On Business, Enterprise and Edu workspaces an admin has to allow it under Permissions & Roles → Connected data.
- Go to Settings → Connectors. The button to create a custom connector only appears once developer mode is on.
- Add the URL above with transport Streamable HTTP and authentication None.
Available on Plus and Pro; on workspace plans an administrator can disable developer mode or allowlist specific connectors.
Claude Code
One command — --transport http is what makes it a remote connector rather than a local process:
claude mcp add --transport http second-chair https://insightbridge-secondchair.aicolab.org/mcpRe-adding after a change needs claude mcp remove second-chair first.
VS Code (GitHub Copilot), Cursor and Windsurf
All three take the same shape of entry. In VS Code run MCP: Open User Configuration from the command palette, or create .vscode/mcp.json in a workspace; in Cursor use Settings → MCP.
{
"servers": {
"second-chair": {
"type": "http",
"url": "https://insightbridge-secondchair.aicolab.org/mcp"
}
}
}Some clients name the top-level key mcpServers rather than servers; if the server does not appear, try the other spelling.
Then ask something like “Using the legal-AI corpus, which claims about AI replacing lawyers are made by vendors with no independent voice present to check them — and what evidence do they actually offer?”
Works with any MCP client. The endpoint speaks JSON-RPC over a stateless one-POST Streamable-HTTP adapter at POST /mcp; there is no SSE stream and no authentication, so a client that insists on OAuth should be set to “no auth”.
How to read what comes back
- Source counts are not prevalence. This is a curated evidence base of 600 published works, not a survey of the profession. "N sources say X" measures the corpus, never the field.
- Positions are relative to each cluster's own proposition — Supports, Builds on, Unclear, Mixed, Redirects, Opposes measure stance-toward-a-framing. 94.6% of the voice lens is Supports/Builds on and there is no Opposes at voice level at all, so agreement is the default and only
contested_topicsfinds real divergence. - Era is ordinal, not a category: Foundations (pre-2015) → Early automation debate (2015-2019) → Pre-generative (2020-2022) → Generative shock (2023-2024) → Adoption at scale (2025) → Current wave (2026). The tools always return era rows in that order.
- Absence of scrutiny is a finding. 53 of 106 clusters carry only one voice, and 13 carry vendor or AI-native voices with no academic, regulatory or civil-society voice present — see
unscrutinised_topics. - topic_id is the topicClusterId, not the source
cluster_id. The two are offset by one, so the wrong one quietly returns the adjacent topic.
Try the tools
Pick a tool, adjust the inputs, and run it against the live server — the exact request an agent would make. Every tool is pre-filled with example values that return real results.
corpus_overviewNo inputs — just run it.
{
"method": "tools/call",
"params": {
"name": "corpus_overview",
"arguments": {}
}
}Run a tool to see its output.