How do I authorize Tempreon in the Kimi Code CLI?

Authorizing is a separate step from adding the server. Once Tempreon is added through /mcp-config, run /mcp-config login tempreon, and a browser tab opens with the Tempreon consent screen to sign in and approve. If you later hit an auth loop or stale tokens, run that login command again, or remove and re-add the Tempreon entry through /mcp-config.

Where is Tempreon's MCP configuration stored in Kimi, and can I edit the file directly?

Kimi Code stores it in ~/.kimi-code/mcp.json for all projects, or .kimi-code/mcp.json inside a project to scope it to one. Manage it from inside Kimi Code with /mcp-config, and use /mcp to check server status. Prefer /mcp-config over hand-editing, since the file location and the command surface have both moved between Kimi versions.

I switched Claude Code to Kimi, but the Tempreon tools disappeared. How do I fix it?

The environment variables only swap the model; the Bridge is configured separately, so changing them doesn't move your Tempreon connection. Confirm Tempreon still shows as connected under /mcp in Claude Code, and restart the session after changing any environment variables. If it isn't listed, add it with claude mcp add and complete the OAuth login.

Kimi Setup (Kimi Code CLI & Claude Code) — Docs

Overview

Kimi is Moonshot AI's model family. The newest flagship, Kimi K3, was announced on July 16, 2026, with open weights released on July 27, 2026 — a 2.8-trillion-parameter model with a roughly one-million-token context window. There are two working routes to pair Kimi with Tempreon: the native Kimi Code CLI, or running Kimi as the model inside Claude Code. Both give you the same result — your AI loads your Core Imprint, searches your Knowledge Vault, and remembers what it learns, across every tool you connect.

Which route to pick: if you already use Claude Code, route B slots Kimi into your existing setup with an environment block. If you want Moonshot's own harness, route A configures it from inside the Kimi Code interface.

Quick path: in Kimi Code, run /mcp-config → add Tempreon by URL → /mcp-config login tempreon → approve in the browser → done.

Prerequisites

A Moonshot account with Kimi Code access (route A) or a Moonshot API key (route B)
A Tempreon account with at least one available Bridge slot
A working default browser for the authorization step

Step-by-Step Setup

Route A: Kimi Code CLI (native)

Kimi Code manages MCP servers from inside its own interface. Start Kimi Code and run:

/mcp-config

That opens the MCP configuration surface, where you can add a server. Use the Tempreon URL:

https://api.tempreon.com/functions/v1/tempreon-mcp/mcp

The URL is also shown in your Bridges dashboard. Then authorize:

/mcp-config login tempreon

A browser tab opens with the Tempreon consent screen — sign in if prompted and approve. /mcp shows server status once it's connected.

Kimi Code stores this configuration in ~/.kimi-code/mcp.json for every project, or .kimi-code/mcp.json inside a project to scope it to one. Moonshot's command surface has changed between versions, so prefer /mcp-config over any CLI subcommand you may find in an older write-up.

Route B: Kimi inside Claude Code

Moonshot runs an Anthropic-compatible endpoint, so Claude Code can drive Kimi directly. Point Claude Code at it:

export ANTHROPIC_BASE_URL=https://api.moonshot.ai/anthropic
export ANTHROPIC_AUTH_TOKEN=your-moonshot-api-key
export ANTHROPIC_MODEL="kimi-k3[1m]"
export ANTHROPIC_DEFAULT_OPUS_MODEL="kimi-k3[1m]"
export ANTHROPIC_DEFAULT_SONNET_MODEL="kimi-k3[1m]"
export ANTHROPIC_DEFAULT_HAIKU_MODEL="kimi-k3[1m]"
export CLAUDE_CODE_AUTO_COMPACT_WINDOW="1048576"

Set the whole block, not just the first two lines — Moonshot documents every model tier explicitly, and a partial configuration tends to fail quietly rather than error. Note that the model id depends on which Moonshot endpoint you're on: the platform endpoint above (api.moonshot.ai/anthropic) uses kimi-k3[1m], while a Kimi Code subscription goes through https://api.kimi.com/coding/ and uses the shorter k3[1m]. Using the other route's id returns a model-not-found error, so match the id to the base URL. Your existing Claude Code MCP setup — including Tempreon — keeps working unchanged, because MCP lives in the harness, not the model. If you haven't connected Tempreon in Claude Code yet:

claude mcp add --transport http tempreon https://api.tempreon.com/functions/v1/tempreon-mcp/mcp

then run /mcp inside Claude Code to complete the OAuth login. The full walkthrough is in the Claude Code guide.

Route C: kimi.com

As of August 2026, the consumer web app at kimi.com has no custom connector path — you can't add Tempreon there. Use route A or B; anything your AI learns through them is waiting in your Tempreon account regardless of which surface you work from.

Troubleshooting

Connection fails immediately. Check the URL ends in /mcp — no trailing slash. The full URL is https://api.tempreon.com/functions/v1/tempreon-mcp/mcp.

Hand-edited config isn't being picked up. Kimi Code reads ~/.kimi-code/mcp.json (all projects) or .kimi-code/mcp.json (one project), and the location has moved between versions. Re-add the server through /mcp-config and confirm with /mcp rather than editing files directly.

Auth loop or stale tokens (route A). Re-run /mcp-config login tempreon for a fresh browser flow. If it still loops, remove the Tempreon entry via /mcp-config and add it again.

Route B connects but behaves oddly. Check that every variable in the block above is set. Moonshot's documented setup names each model tier, and a partial environment can leave Claude Code pointed at a model your key doesn't serve.

Kimi responds but Tempreon tools are missing (route B). The env vars only swap the model — the Bridge is configured separately. Confirm Tempreon shows as connected under /mcp in Claude Code, and restart the session after changing environment variables.

Also connect: GLM · OpenCode · Qwen Code

Try saying...

"Start my session." — triggers session_start to load your Core Imprint
"Remember: we deploy Fridays only after the staging soak passes." — stores a working rule your AI keeps across models
"Search my knowledge base for notes on the billing refactor." — exercises retrieval end-to-end