OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Evaluation & Observability

Raindrop

Workshop is a local developer tool for inspecting, replaying and debugging AI-agent runs. It streams tokens, tool calls and decisions, and supports creating and running evaluations against agent behavior.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Raindrop — OpenViST project profile in Agent Evaluation & Observability.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Workshop when you need a local view of an agent run, want to examine a failure through replay, or are building evaluations to check agent behavior.

Who it is for

  • Developers investigating AI-agent behavior and failures.
  • Teams using coding agents in the listed ecosystems.

Documented uses

  • Inspect an agent run’s tokens, tool calls and decisions.
  • Replay a run while diagnosing an agent failure.
  • Create and run evaluations based on agent behavior.
Before you choose

Requirements & limits

Workshop is described as free and open source. Raindrop for Good is a separate application-based program: qualifying teams building agents that help people receive the broader platform for free.

  • The listed ecosystems are Claude Code, Codex, Devin, Cursor and OpenCode.
  • Workshop focuses on local debugging and evaluations; the separate Raindrop for Good program concerns the broader platform.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Apply for Raindrop for Good

Submit an application for free Raindrop access for an agent whose main job is helping people.

Before you start

  • An agent focused on helping people; eligible applicants may include nonprofits, public-benefit companies, government or civic projects, mutual aid groups, and researchers. Registered nonprofit status is not required.
  • Information about what the agent does, who it helps, and its current stage.

Steps

  1. Open the Raindrop for Good application.

  2. Provide your name, email, and organization; website is optional.

  3. Choose the agent’s current stage and describe what it does and whom it helps.

  4. Submit the application and wait for a response; Raindrop says its team responds within a week.

Expected outcome

If accepted, Raindrop says it will set up a free account and help connect the first traces.

Guide limitations

  • Free access is for qualifying teams and continues while the work continues; Raindrop says it will discuss substantially increased usage first.
  • Applying does not guarantee acceptance.
  • Raindrop says its platform includes PII redaction in its SDKs and on its servers and is SOC 2 Type II certified.
Documentation 1 ↗
Evidence and transparency

Sources & review

raindrop.ai ↗Guide documentation ↗
Source record details
Source availability
Active
Sources checked
4 November 2026
Original website title
Workshop | Raindrop - Debug your AI agent locally · Official homepage title · retained in its source language
How we review sourcesSuggest a correction