# Write social media posts directly in Kimi Code

> Source: https://tareno.co/agents/kimi
> Language: en
> Add Tareno to the Kimi Code CLI as a remote HTTP MCP server with OAuth, turn changelogs and explanations into X and LinkedIn drafts and approve every request in Tareno.

Kimi Code is Moonshot AI's command-line coding agent. Its MCP client can reach remote servers over HTTP, including servers that sign you in with OAuth, which is the route Tareno uses. With Tareno added, a session that has just read a changelog or untangled a tricky bug can save a post about it as a draft, while every scheduling and publishing request waits for your approval in Tareno. This page covers the Kimi Code CLI only. It does not make claims about any Kimi model, and Tareno has not completed a client test, so the setup follows Moonshot's documentation.

## Connection facts

| Field | Value |
| --- | --- |
| Product surface | Kimi Code (kimi) |
| Available connection | Custom MCP connection |
| Connection evidence | documented |
| Methods | mcp |
| Authentication | oauth |
| Setup | [Setup guide](https://tareno.co/docs/mcp) |
| Install or connect | [Open setup](https://tareno.co/docs/mcp) |
| Evidence observed | 2026-10-25 |
| Facts reviewed | 2026-10-25 |

Publication status and a working connection are separate facts. Evidence for one product surface does not establish installation or compatibility in another surface.

## How do I connect Tareno to Kimi Code?

Run the kimi mcp add command with the http transport, OAuth as the authentication, a name such as tareno and Tareno's MCP address. Then run kimi mcp auth tareno to complete the browser sign-in, and kimi mcp test tareno to check it. Servers are stored in ~/.kimi/mcp.json. The steps follow Moonshot's documentation, and Tareno has not completed a client test.

## Connect Kimi Code in four steps

1. **Add the server.** Moonshot's documentation gives the form kimi mcp add with the http transport, followed by a name and the URL. For a server that signs you in, add the OAuth authentication option. Use the address shown on the right for Tareno.
2. **Complete the sign-in.** Run kimi mcp auth with the server name. A browser opens for the authorization flow, and the token is saved under ~/.kimi/mcp-oauth/ for later use. Sign in with your own Tareno account, not a shared one.
3. **Test the server.** kimi mcp test with the server name checks that it responds, and kimi mcp list shows what is configured. The documentation also says servers live in ~/.kimi/mcp.json, where a remote server uses a url field and optional headers.
4. **Run a read-only call.** Ask Kimi Code to call `list_accounts`. The account you meant to post from appearing in the answer is the confirmation; an empty answer means that account still has to be connected inside Tareno.

Tareno's remote MCP endpoint:

```json
"url": "https://tareno.co/api/mcp"
```

## Where Kimi Code helps with Tareno

A coding agent that has just read the changes is well placed to describe them. These jobs use that, and each has a workflow guide.

### A changelog becomes a short X post

Changelogs are written for people who already use the tool and read poorly as posts. Kimi Code can take the entries, pick the one that changes what a user does and write a post within 280 characters that links the full list, leaving the other entries for the next post you decide to write.

### A hard bug becomes a LinkedIn explanation

After a long debugging session the reasoning is fresh: what looked wrong, what was wrong, how it was confirmed. Kimi Code can turn that into a LinkedIn explanation for other engineers, with the stack trace and any secrets removed, so a lesson that would be forgotten by Friday becomes a draft.

### Skills and hooks around the connection

Moonshot's documentation lists AGENTS.md memory, permission modes, hooks and skills for the CLI. A team can keep its rules for public posts, such as no customer names and a limitation in every case study, in AGENTS.md so that every session starts from them.

## Prompts to copy

### Check the connection

```text
Use Tareno to list my connected social accounts with platform and account name. Read only: do not create, schedule or publish anything. Tell me which account you would use for a changelog post on X and which for a LinkedIn explanation.
```

### Changelog to X

```text
Read CHANGELOG.md for the newest version and pick the one entry that changes what a user does. Write an X post within 280 characters about it, with the version number and no hashtags, and tell me which entry you chose and which you left out.
```

### Bug to explanation

```text
Summarize the bug we just fixed for other engineers in a LinkedIn draft of about 200 words: the symptom, the real cause, how we confirmed it and one thing to check in similar code. Remove stack traces, file paths with usernames and any key or token.
```

### Save the drafts

```text
Use Tareno to confirm my X account [handle] and my LinkedIn profile [name], then save the two approved texts as separate drafts. Return each draft ID with its text, and request no time until I give you the day and the time zone.
```

### Request the slots

```text
Take the drafts [draft IDs] and ask Tareno to schedule them for [Wednesday 10:00] in [Asia/Singapore]. Return the action IDs and remind me that each request still waits for my approval in the Tareno dashboard.
```

## Guides by network

- LinkedIn: [Can Kimi Code turn a debugging session into a LinkedIn explanation?](https://tareno.co/agents/kimi/linkedin) ([Markdown](https://tareno.co/agents/kimi/linkedin.md))
- X: [Can Kimi Code turn a changelog entry into an X post users will read?](https://tareno.co/agents/kimi/x) ([Markdown](https://tareno.co/agents/kimi/x.md))

## Good to know before you start

- Tareno has not completed a client test with Kimi Code. The setup follows Moonshot's documentation, so verify it with a read-only call.
- This page covers the Kimi Code CLI. Naming and versions of Kimi models vary between sources, so it makes no claim about any model.
- Moonshot's documentation says OAuth tokens from an older version are not migrated after an upgrade, so run kimi mcp auth again if the server shows as needing authorization.
- Static API keys passed as headers are stored in the config file as written. Use the OAuth sign-in for Tareno instead of pasting a key.
- Scheduling, publishing, edits to scheduled posts and deletions create approval actions in Tareno. Kimi Code can request them but cannot approve them.

## Kimi Code compared with other CLI agents

Codex, Claude Code and Grok Build also run in the terminal and reach Tareno as remote MCP servers. Kimi Code's own commands for adding, authenticating and testing a server are the main difference, along with its AGENTS.md and skills support. Choose it if it is already your daily tool, and keep a read-only check in your routine either way.

## Frequently asked questions

### Which Kimi product does this page cover?

The Kimi Code CLI, Moonshot AI's command-line coding agent. It makes no statement about the Kimi chat app or about any specific model, because the documentation it relies on describes the CLI's MCP client rather than model capabilities.

### Where does Kimi Code store MCP servers?

In ~/.kimi/mcp.json, according to Moonshot's documentation. A remote server uses a url field, with an optional headers object for values such as API keys. OAuth tokens are kept separately under ~/.kimi/mcp-oauth/ after you authorize.

### How do I sign in to Tareno from Kimi Code?

Add the server with OAuth as the authentication, then run kimi mcp auth with the server name. A browser opens, you sign in with your own Tareno account, and Kimi Code saves the token so later sessions do not ask again.

### How do I know the server works?

Run kimi mcp test with the server name, then ask for `list_accounts` in a session. The first checks the transport and the second confirms that Tareno returns your accounts, which is the check that matters before any draft is saved.

### Is the Kimi Code setup proven by Tareno?

Not yet. The steps follow Moonshot's documentation, and Tareno's OAuth is standard, but a full read-only run through Kimi Code has not been completed. Treat your first `list_accounts` call as the test for your own setup.

### Does Kimi Code get to press publish?

No. A scheduling or publishing request becomes an approval action in Tareno, and nothing reaches X or LinkedIn until you approve it in the dashboard. Kimi Code can only report that the action is pending and read its status afterwards.

### Should I use an API key header instead of OAuth?

Prefer OAuth. A key placed in a header is stored in the config file as written and can end up in backups or screenshots. The sign-in flow keeps the credential in Kimi's token store and ties it to your own account.

### What will Kimi Code ask Tareno for first?

Usually `list_accounts` to confirm the destination, then `create_drafts` to save the text. `schedule_posts` comes last and returns an approval action instead of publishing anything, and `get_action_status` tells you what became of the request.

## Tareno tools used

- `list_accounts`
- `create_drafts`
- `schedule_posts`
- `get_action_status`

Approval model: Tareno saves drafts directly. Scheduling, publishing, edits to scheduled posts and deletions become pending actions that a person approves in the Tareno dashboard; no agent can approve them.

## Related

- [A changelog post on X](https://tareno.co/agents/kimi/x)
- [A LinkedIn explanation of a fixed bug](https://tareno.co/agents/kimi/linkedin)
- [Remote MCP setup](https://tareno.co/docs/mcp)
- [Claude Code × Tareno](https://tareno.co/agents/claude-code)
- [Grok Build × Tareno](https://tareno.co/agents/grok-build)
- [Approval workflows](https://tareno.co/features/approval-workflows)
