Citadel Predict
Quiet signalStop AI agents from blowing your token budget
Launched 5d ago
Signal
18
Velocity
21
What this means
+428%Launching in a 428% WoW growing category.
Open Source had 132 launches this week vs 25 last.
20Founder Conviction Index: 20 — low signal.
Few of the conviction sub-signals (reputation, velocity, buyer-intent, tagline clarity) are firing yet.
Prediction
Top-5 finish probability
13%
today
Projected end-of-day votes
2range 2–3
Trajectory
stable
Not enough snapshots yet to detect trajectory.
Speed vs peers
0.8×
156 comparable launches
About
Citadel Predict forecasts AI agent token usage and cost before the agent runs — not after, like existing tools (Langfuse, Helicone, LangSmith). It analyzes your task and tool list, then uses a calibrated ML model to give you an expected token range, confidence level, and out-of-distribution risk flag — now converted into an estimated USD cost too. Ships as a pip package + CLI, an MCP server for Claude Desktop/Code, and a hosted API. Tested on Linux, Windows, and macOS.
Attention & discussion velocity
Not enough snapshots yet. Run the snapshot worker every 15 min to build the time-series.
Performance
Pace needs two or more snapshots. This launch has 1, so velocity and momentum read as unmeasured rather than zero.
Velocity—
not measured
Momentum—
not measured
Virality—
not measured
Engagement—
not measured
Signal sources
1 of 6 active- Launch tracking1 reading since we picked it up
- AI analysisnot yet enriched
- Comment intelligencethread not yet analysed
- Outside discussionnot checked — domain is a storefront or shared
- Revenue signalsno match on this domain
- Editorial scoringbelow the signal threshold for this pass
A source that found nothing is a measurement. A source that has not run is a gap. Neither means the launch lacks the thing.