Guides

Each guide gets one thing done with Kindgi: pick an area, then the task.

A guide assumes you have done Start. Its code is in tabs, one per SDK the guide covers. Every guide was run as written against the version of Kindgi these docs describe.

• Tools: write a tool in TypeScript or Python, call an HTTP API without code, give a tool a secret, mark it read-only, and use an MCP server's tools.
• Agents: write an agent, give it input, get a typed answer, choose its model, hold a conversation and cap what a turn may spend.
• Models: connect Anthropic, Gemini on Vertex AI, an OpenAI-compatible endpoint, or a model you serve yourself.
• Flows: pass data between steps, branch, loop, run steps in parallel, retry, wait for a person, and dry-run a flow.
• Runs: start runs from the CLI or your app, retry safely, follow them live (from the browser too), read their journal, cancel and list them.
• Orgs and projects: give each customer an org with its projects, and run and read their work by project or org.
• Webhooks: get a signed run.finished request when a run ends, verify it in your app, and test and replay deliveries.
• Guardrails: check an agent's answers, give a check its settings, and choose whether a failure stops the turn.
• Approvals: have a person approve what an agent does before it happens.
• Secrets and env: where a pack's settings and secrets live on your machine and in a deployment, how your code gets them, and the sandbox kindgi dev runs it in.
• Sign-in: sign in to the console, keep a session, and see the enterprise sign-in options Kindgi builds with you.
• Cost and provenance: trace an answer to the model and tool calls behind it, and what each call cost.
• Evals: judge runs' answers, build a test set from them, change a prompt or a setting without a deploy, compare the new version on the test set, and release it. You can also propose a change through review, or let a pass look for a better setting or prompt.