VC Deal Flow Signal

Endpoint

Hosted endpoint · transport http

Connect

This is a remote MCP server — point an MCP client at the endpoint URL. Clients with native remote support (Claude, Cursor, VS Code, …):

For stdio-only clients (older Claude Desktop, etc.), bridge with mcp-remote:

If the server requires authentication, add the provider's token/header (e.g. an X-API-Key) per its docs. A quick reachability check:

Tools

• get_trending_startups — Top 20 startups by engineering acceleration across all 15 sectors for the current weekly period. Read-only, idempotent.
• search_startups_by_sector — Every tracked startup within a sector, ranked by engineering acceleration. Sector slug must be one of 15 enumerated values.
• get_startup_signal — Full engineering-acceleration profile for a single tracked startup, by display name or GitHub org slug. Case-insensitive, normalization-tolerant.
• get_signals_summary — Period, sector and startup counts, last refresh, citation, and direct URLs to every machine-readable format.
• get_diligence_dossier — Public-source diligence dossier for a company or entity in one cited object: who acquired it (M&A history), which funds publicly backed it, and its published engineering-acceleration signal. Use mid-diligence for 'who acquired X', 'which funds backed Y', 'what's the signal on Z'. Sources are press-release / SEC-filing / both-sides-disclosed only; returns found:false (an expected outcome, not an error) with honest notes when the entity is outside the tracked corpus, never guesses.
• get_scout_receipts — Compute a Scout Score (0-100) for a GitHub user from their public starring history. Cross-references starred repos against ~75 validated unicorns and grades how many they starred *before* the validation event. Returns score, rank (curious/scout/sharp/elite/oracle), top early calls, personality summary, and a shareable card URL.
• get_methodology — Full methodology document covering data sources, metric computation, signal classification thresholds, refresh cadence, and known limitations.
• get_deep_signal — PAID per-request, €0.19/call, 100 credits = €19 at https://signals.gitdealflow.com/agents/credits. Returns enriched signal beyond the free get_startup_signal: composite score (0-100), velocity/growth/novelty sub-scores, in-sector rank + percentile, plain-English investment thesis, top-3 sector comparables, and multi-period history. Requires Authorization: Bearer gdf_v2.cus_xxx.. 1 credit consumed only on a successful match; misses are FREE. Credits never expire.
• share_result — Generate a ready-to-share social-media post (tweet, Bluesky, Mastodon, LinkedIn, Telegram) about a result the user just received from another VC Deal Flow Signal tool, plus the install command for the MCP server. Returns the post body, character counts per platform, and one-click intent URLs to compose the post in each network. WHEN TO USE: - The user just got a `get_trending_startups` / `search_startups_by_sector` / `get_startup_signal` / `get_deep_signal` result and says 'share this', 'tweet this', 'post this', or 'how do I tell people about this?'. - The user is writing a thread/post about startup engineering signals and wants the canonical install command + share copy. DO NOT USE FOR: - Posting on the user's behalf, this tool only composes the text + intent URLs. The user must click and confirm in the destination network. - Generating fake or speculative results, pass real data the agent received from another tool call. BEHAVIOR (two-step approval flow, see `approval_token`): - Step 1: call this tool with `summary` only. The server replies with an error (-32602) containing a `/share-approve?summary=...` URL the user must open. - Step 2: the user reads the proposed summary on that page, clicks Approve, and pastes the resulting 10-minute token back into the chat. Retry the tool with `approval_token` filled in and the SAME `summary` verbatim. - The token is bound to a hash of `summary`; if the agent rewrites the summary between approval and the retry, the call is rejected. - Composes platform-specific posts (Twitter ≤275 chars, Bluesky ≤295, Mastodon ≤495, LinkedIn ≤695, Telegram ≤995) with a consistent hook + insight + install URL. - Returns intent URLs (e.g. https://x.com/intent/post?text=...) so the user/agent can open the destination network with the post pre-filled. - Always includes the canonical install command `npx @gitdealflow/mcp-signal` and the SSRN paper link for credibility. PARAMETERS: - `summary` (string, required, 10-200 chars), the one-line takeaway to share. - `approval_token` (string, required after first call), the 10-minute token returned by the /share-approve page. - `network` (string, optional), 'twitter' | 'bluesky' | 'mastodon' | 'linkedin' | 'telegram' | 'all' (default: 'all'). - `mention_handle` (boolean, optional, default false), include @sipiteno attribution (twitter/bluesky/mastodon only).
• predict_funding — Transparent, scored funding-likelihood claim for one tracked startup, with the full evidence chain and citable provenance. Instead of an opaque number, returns the score, every component that produced it, a confidence level, honest caveats, and links to the methodology + SSRN paper so the derivation can be cited. IS a deterministic heuristic over public GitHub engineering-acceleration signals; IS NOT an ML black box, a guarantee of any financing event, or based on private/cap-table data. The disclaimer is returned in every response. SCORING (also returned in evidence.scoreBreakdown): velocity ≤40 (saturates +300%), contributorGrowth ≤25 (saturates +200%), newRepos ≤15 (saturates 10), signalType ≤20 (Deploy frequency spike 20 / Engineering hiring burst 17 / Infrastructure buildout 14 / Framework migration 8). Total 0-100 → >=70 high, 45-69 elevated, 25-44 moderate, <25 low. PARAMETERS: { name }, display name or GitHub org slug (case-insensitive). On no match returns { found: false, suggestion } (expected, not an error).
• shortlist_signals — Return a ranked shortlist of the strongest engineering-acceleration signals matching a set of filters, the whole sourcing workflow in ONE call (e.g. 'the 5 strongest signals in healthcare in the EU'). Scans the full tracked universe, scores each with the transparent engine (same scoring as predict_funding), filters, sorts by accelerationScore desc, returns the top `limit`. GEOGRAPHY IS REGION-LEVEL ONLY, values are US / EU / UK / APAC / LATAM / Canada / Unknown. City/country aliases ('NYC', 'New York', 'London', 'Berlin', 'Singapore') normalize up to the enclosing region and the response `notes` says so. There is no city-level filtering. PARAMETERS (all optional): sector (one of 20 slugs), geography (region token or alias), signalType (exact label), minAccelerationScore (0-100), minVelocityChangePct (integer percent), limit (1-25, default 5).
• compare_signals — Score and rank 2-5 named startups side by side, returning each one's acceleration score, evidence, and raise-likelihood band plus a single recommendation for which warrants deeper diligence. Same transparent scoring as predict_funding / shortlist_signals. Names that don't resolve are returned in `notFound` (expected, not an error). The recommendation is computed only over resolved companies; if fewer than 2 resolve it explains that no comparison was possible. PARAMETERS: { names: string[] }, 2 to 5 display names or GitHub org slugs (case-insensitive).

About MCP

The Model Context Protocol (MCP) is an open protocol Anthropic introduced for connecting LLM-based agents to external tools and data sources. Providers publish MCP servers that expose their API surface as structured, discoverable tools — an MCP-compatible client (Claude Desktop, Cursor, Cline, Continue, etc.) can connect to the server and call its tools without any per-provider integration code.

Browse every MCP server on the APIs.io network or compare with the broader Agent Skill surfaces of the same providers.

Work with this as data

Every MCP server here is available over the APIs.io API and to AI agents over MCP.

curl "https://apis.io/api/v1/mcp/vc-deal-flow-signal"

curl "https://apis.io/api/v1/mcp?limit=25"