Dated, attributed founder and investor positions — the disagreement, with timestamped receipts.
Scored across 5 tools
Each tool has a distinct retrieval mode: routing dilemmas to debates, browsing the registry, listing complete stances, pulling one claim's context, and tracing a speaker's evolution. Where outputs overlap, descriptions explicitly disambiguate them, such as find_perspectives returning a sample versus get_stances returning the complete set.
All five tools follow a consistent lowercase snake_case verb_noun pattern: find_perspectives, get_context, get_stances, list_debates, trace_evolution. The verbs clearly describe the action and the nouns clearly identify the object.
Five tools is well-scoped for a read-only perspective graph server. Each tool earns its place by covering a distinct access path without redundancy or bloat.
The tool set covers the full read path: discover debates, route a dilemma to relevant debates, retrieve complete stances, inspect a claim in context, and track a speaker's evolution over time. For a read-only knowledge graph, there are no obvious dead ends or missing operations that would block a founder's workflow.
Route a founder's dilemma to the debates that genuinely bear on it. Present the readings SEPARATELY — never merge them into one answer. Respect the question's altitude: scale-up material does not answer a pre-PMF dilemma. When coverage is thin it says so — advise from general knowledge rather than stretching a weak match. Returns ≤4 debates and ≤36 stances — a sample, not the full set; get_stances(debate_id) has a debate's complete stances, each with speaker, date, strength and a receipt.
Inputs: topic (required).
The transcript around one claim — the moment in context, with its receipt. Spoken: the episode's public transcript, windowed ±90s around the claim's timestamp. Written (an essay): the verbatim excerpt snapshotted at digest plus a link into the original; never the surrounding text, never a live fetch. Quote the receipt — episode + timestamp, or essay link + excerpt — so the founder can check the moment.
Inputs: claim_id (required).
Every stance on one debate, grouped by reading — raw material for your own who-would-push-back read. No synthesis. Attribute and date every position (dates are episode publication dates); keep hedged claims hedged. The graph never says who opposes whom — work out yourself which positions push back on the founder's leaning, and name them.
Inputs: debate_id (required).
Browse the debate registry: id, question, readings, stance count, born (earliest publication date). Optional query: a debate matches when its question has every word of it, case-insensitive, word forms folded (hire/hiring, co-founder/cofounder). Browsing only — it does not rank against a question; use find_perspectives for that. Defaults to debates with 2+ stances, 120 rows a page — page with offset.
Inputs: limit, query, offset, min_claims.
One speaker's takes in order of episode publication date (optionally on one debate) — reversals and re-assertions visible, never judged. Attribute and date every position (dates are episode publication dates); keep hedged claims hedged. A 2023 position is a 2023 position; if the same person later said otherwise, show both.
Inputs: person (required), debate_id.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Where informed investors disagree, the evidence each side stands on, and the date that settles it.
Startup investors, funding rounds and finance reports, with page-cited evidence for your AI.
Dated, sourced facts about AI and tech companies: what was true on a day, and what changed since.
Dated record of what companies run, sell, hire and say on their websites, with receipts.
Add one secure layer between your agents and this server.