> For the complete documentation index, see [llms.txt](https://docs.flora.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.flora.ai/mcp/retired-tools.md).

# Retired FLORA MCP tools: execute and older names

Move off retired FLORA MCP tools: replace execute, flora\_generate, flora\_get\_run, get\_more\_tools and search\_docs with the current tools, and refresh a client's cached tool list.

[FLORA MCP](https://flora.ai/mcp) has retired five tool names. The server no longer lists them, and a call to any of them returns an error that names the replacement. If your agent, script or saved prompt still uses one, this page shows what to call instead and how to make your client forget the old list.

## Old name to new name

| Retired name     | Replacement                                                  | What to change                                                                                            | Retired    |
| ---------------- | ------------------------------------------------------------ | --------------------------------------------------------------------------------------------------------- | ---------- |
| `execute`        | The named `flora_*` tools. Find them with `flora_list_tools` | Call the tool for each operation directly. For scripts, use the [FLORA API](https://docs.flora.ai/api.md) | 2026-10-08 |
| `flora_generate` | `flora_create_generations`                                   | Wrap the same arguments as one item in `generations: [...]`                                               | 2026-09-11 |
| `flora_get_run`  | `flora_list_generations`                                     | Pass `run_ids: [run_id]`, plus `technique_id` for Technique runs                                          | 2026-09-11 |
| `get_more_tools` | `flora_list_tools`                                           | Call `flora_list_tools` with no arguments                                                                 | 2026-10-03 |
| `search_docs`    | `flora_discover_skills`                                      | Call `flora_discover_skills` for workflow guidance. Tool descriptions document each tool                  | 2026-08-27 |

## The error a retired name returns

A call to a retired name doesn't run anything. It returns a tool error (`isError: true`) whose text is this JSON:

```json
{
  "ok": false,
  "error": "execute was retired: this server no longer runs SDK code. Call flora_list_tools({}) to find the dedicated flora_* tool for the operation and call it directly; independent calls can run concurrently. For workflow instructions call flora_discover_skills. Refresh your tool list (tools/list); this server no longer registers execute.",
  "replacement": "flora_list_tools",
  "retired": "execute"
}
```

Read `replacement` and call that tool. The `error` text for each name:

| `retired`        | `error`                                                                                                                                                                                                                                                                                                                                                                                                         |
| ---------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `flora_generate` | flora\_generate was replaced by flora\_create\_generations. Wrap the same arguments as one item: flora\_create\_generations({ generations: \[{ workspace\_id, project\_id, type, prompt, model?, params?, reference\_node\_ids? }] }). Poll the returned run\_id with flora\_list\_generations({ run\_ids: \[run\_id] }). Refresh your tool list (tools/list); this server no longer registers flora\_generate. |
| `flora_get_run`  | flora\_get\_run was replaced by flora\_list\_generations. Poll with flora\_list\_generations({ run\_ids: \[run\_id] }) — pass technique\_id too for flora\_run\_technique runs. Read every run\_id of a batch in one call. Refresh your tool list (tools/list); this server no longer registers flora\_get\_run.                                                                                                |
| `get_more_tools` | get\_more\_tools was replaced by flora\_list\_tools. Call flora\_list\_tools({}) to discover the tools available on this connection. Refresh tools/list for complete descriptions and schemas. …                                                                                                                                                                                                                |
| `search_docs`    | There is no search\_docs tool on this server. For workflow guidance call flora\_discover\_skills; the server instructions and tool descriptions are the reference for tool usage. Refresh your tool list (tools/list).                                                                                                                                                                                          |

## `flora_generate` to `flora_create_generations`

`flora_create_generations` starts 1 to 20 generations in one call, so the request is an array even for one item.

Before:

```json
{
  "workspace_id": "ws_123",
  "project_id": "prj_abc",
  "type": "image",
  "prompt": "A ceramic pour-over dripper on a warm grey backdrop",
  "model": "t2i-gemini-3.1-flash-image"
}
```

After:

```json
{
  "generations": [
    {
      "project_id": "prj_abc",
      "type": "image",
      "prompt": "A ceramic pour-over dripper on a warm grey backdrop",
      "model": "t2i-gemini-3.1-flash-image"
    }
  ]
}
```

`workspace_id` is now optional: FLORA bills the workspace the project belongs to. The response is a list in input order, and each entry carries `index` and `ok`:

```json
{
  "generations": [
    {
      "run_id": "run_a1",
      "type": "generation",
      "model": { "model_id": "t2i-gemini-3.1-flash-image" },
      "estimated_seconds": 65,
      "charged_cost": 0.072,
      "project_id": "prj_abc",
      "canvas_url": "https://app.flora.ai/projects/…",
      "poll_url": "https://app.flora.ai/api/v1/runs/run_a1",
      "node_id": "7f6ae6da-4a0e-4a2f-9c6a-2c1c6b8d1f21",
      "index": 0,
      "ok": true
    }
  ]
}
```

Two things are new. Add `"quote": true` to price the request in US dollars without starting it (see [Cost, balance and limits](https://docs.flora.ai/mcp/cost-and-limits.md#price-a-request-with-quote-true)). And when one item in a batch fails, the others still run, so never retry the whole batch.

## `flora_get_run` to `flora_list_generations`

`flora_list_generations` reads up to 20 runs per call, so a batch is one call instead of one per run.

Before:

```json
{ "run_id": "run_a1" }
```

After:

```json
{ "run_ids": ["run_a1"] }
```

For a Technique run, keep passing `technique_id`:

```json
{ "run_ids": ["run_t1"], "technique_id": "tech_abc" }
```

The response wraps the runs in a list, one entry per ID: `{"runs": [{"run_id": "run_a1", "status": "running", …}]}`. `flora_list_generations` also accepts `run_id` for a single run, but `run_ids` is the form to use. Polling now has a minimum interval; see [Runs](https://docs.flora.ai/mcp/runs.md#poll-a-batch).

## `execute` to the named tools

`execute` ran SDK code in a sandbox on the server. It was deprecated in favor of the named tools and retired on 2026-10-08. The server no longer runs code. Every operation it was used for now has its own tool:

| What the `execute` code did               | Tool to call                                                                               |
| ----------------------------------------- | ------------------------------------------------------------------------------------------ |
| Start generations                         | `flora_create_generations`                                                                 |
| Read runs, or list generation history     | `flora_list_generations`                                                                   |
| List, read or run Techniques              | `flora_list_techniques`, `flora_get_technique`, `flora_run_technique`                      |
| List, read, create or rename projects     | `flora_list_projects`, `flora_get_project`, `flora_create_project`, `flora_rename_project` |
| Read the canvas                           | `flora_get_canvas`, `flora_list_canvas_nodes`                                              |
| Add, change or delete nodes and edges     | `flora_add_to_canvas`, `flora_update_canvas_nodes`, `flora_remove_from_canvas`             |
| Lay out the canvas                        | `flora_arrange_canvas`, `flora_organize_canvas_fields`                                     |
| Run nodes already on the canvas           | `flora_run_canvas_nodes`                                                                   |
| Read or write a timeline or deck document | `flora_get_canvas_node_document`, `flora_update_canvas_node_document`                      |
| Upload files                              | `flora_create_asset`, then `flora_complete_asset`                                          |
| List models and their parameters          | `flora_list_models`                                                                        |
| Read balance and spend                    | `flora_get_usage`                                                                          |

Independent calls can run at the same time, so a script's loop usually becomes a few parallel tool calls. For a multi-step job, ask the agent to load a skill with `flora_discover_skills` (see [Skills and plugins](https://docs.flora.ai/mcp/skills.md)). The full list is in the [tools reference](https://docs.flora.ai/mcp/tools.md).

### If you scripted the SDK inside `execute`

Move the script out of the agent and run it yourself against the [FLORA API](https://docs.flora.ai/api.md), with the TypeScript SDK (`@flora-ai/flora`) or plain HTTP. The API takes the same operations the old sandbox did, authenticated with a FLORA API key, and nothing is lost by running it outside the MCP server. Use MCP when an agent should decide what to do; use the API when your code already knows.

## Refresh the tool list in your client

Clients load the tool list when they connect and can keep using it after FLORA changes. If your agent still calls an old name, or doesn't see a new tool, refresh the list:

| Client                   | How to refresh                                                                                                                                                |
| ------------------------ | ------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| Claude (web and Desktop) | Start a new conversation. If the old names persist, open **Settings → Connectors**, disconnect FLORA and connect it again                                     |
| Claude Code              | Run `/mcp`, select `flora` and reconnect, or restart Claude Code                                                                                              |
| ChatGPT                  | Start a new chat. If the old names persist, disconnect and reconnect the FLORA app (see [Use FLORA in ChatGPT](https://docs.flora.ai/mcp/install/chatgpt.md)) |
| Codex                    | Start a new Codex session                                                                                                                                     |
| Cursor                   | In **Cursor Settings → MCP**, turn `flora` off and on again                                                                                                   |
| VS Code                  | Run **MCP: List Servers** from the Command Palette, select `flora` and restart it                                                                             |
| Any other client         | Reconnect the server or restart the client, so it calls `tools/list` again                                                                                    |

To check what the connection offers right now, ask the agent to call `flora_list_tools`. If an old name keeps coming back after a refresh, it's probably written into a saved prompt, a skill or a script. Update it there.

## Related

* [Changelog](https://docs.flora.ai/mcp/changelog.md)
* [Tools reference](https://docs.flora.ai/mcp/tools.md)
* [Troubleshooting](https://docs.flora.ai/mcp/troubleshooting.md#tools-and-tool-lists)


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.flora.ai/mcp/retired-tools.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
