The signal data layer for Yalc. Best when intent (jobs, news, financing, technographics) drives outbound timing rather than static firmographics.
Predictleads provides structured signal data on 100 million plus companies: news events across 29 categories, job openings, technographics (1.2 billion technology detections all time), key customers / supply chain relationships, financing rounds, and similar company lookups. The pitch is "intelligence for AI agents," and the access layer reflects that: REST API, native MCP server, webhooks for real time signals, plus flat file delivery for batch users.
For Yalc workflows, Predictleads is the canonical signal source. Where Crustdata gives you the database (who exists, where), Predictleads gives you the trigger (what's happening, when). When a Yalc agent says "alert me when ICP accounts post a hiring signal" or "trigger outbound when a target company raises Series B," Predictleads is the layer that supplies the event.
Predictleads sits at the **listen** node in Yalc's GTM topology. It's where event driven outbound starts. Most static prospecting (firmographics, ICP filtering) happens upstream in Crustdata. Predictleads kicks in when the question shifts from "who matches" to "who just changed."
The signal listener. Yalc subscribes to Predictleads webhooks on target accounts, classifies the event via Claude, decides whether to trigger an outbound action (sequence, alert, CRM update). The downstream is the standard send pipeline.
Copy paste prompts for Claude Code that invoke Predictleads.
Predictleads ships a native MCP server, which means Claude Code can query it directly during a Yalc session. No first party Yalc skill exists yet, but the MCP integration covers most use cases without one. A future Yalc skill would consolidate webhook setup and signal scoring into a single verb.
→ Request a Yalc skill for this toolPredictleads runs a free tier with 100 API requests per month. Above that, pricing scales with request volume and is custom. The free tier is generous enough to build a real Yalc workflow that triggers on signal events without paying anything for low frequency use cases.
The structure of the API (REST plus MCP plus webhooks plus flat files) supports both interactive Yalc usage (Claude calls Predictleads on demand) and batch usage (download flat files into your own warehouse). Pick the pattern that fits your workflow.
100 API requests per month. Right for piloting and low frequency signal monitoring.
Volume based. Talk to sales for current rates.
Flat file delivery, dedicated infra, SLA. For data teams running batch.
They compose rather than compete. Crustdata is a database (who exists, where, what they do). Predictleads is a signal layer (what just changed). Most Yalc workflows use both, Crustdata to filter ICP, Predictleads to trigger on events.
29 event categories. The high signal ones include financing rounds, executive hires, expansion moves, technology adoption, hiring patterns, layoff news, and product launches. Lower signal events include routine press releases and conference participation.
Yes. Webhooks are first class. You subscribe to event types and target accounts, Predictleads POSTs to your endpoint when it detects matching events. Yalc workflows route the POST through Claude for classification before any action.
Yes. 100 API requests per month. Enough to pilot and to run low frequency monitoring on a small target list.
Most events surface within hours of detection. Some (job openings, news) are near real time. Some (financing rounds) lag because the underlying source (regulatory filings, press) lags.
It's company centric. For person level, pair with Crustdata or Apollo. The strength is company events and technographics.
Or fork the repo and contribute one.