Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion .claude-plugin/marketplace.json
Original file line number Diff line number Diff line change
Expand Up @@ -7,7 +7,7 @@
},
"metadata": {
"description": "Altimate's Claude Code plugin — data tools through the altimate-code MCP server, or the model-free CLI where MCP is disallowed, plus delegation of long dbt jobs to the CLI agent.",
"version": "1.8.0"
"version": "1.8.1"
},
"plugins": [
{
Expand Down
22 changes: 14 additions & 8 deletions plugins/altimate-code/.claude-plugin/plugin.json
Original file line number Diff line number Diff line change
@@ -1,7 +1,7 @@
{
"name": "altimate-code",
"displayName": "Altimate Code",
"version": "1.8.0",
"version": "1.8.1",
"description": "Gives Claude Code altimate-code's model-free data tools (column-level lineage, impact analysis, PII detection, data-diff, FinOps, query equivalence, dialect translation, warehouse queries) through its MCP server, or through `altimate tool run` when MCP servers are not allowed, with delegation to the altimate-code agent for long dbt jobs, access to a linked Altimate workspace's knowledge, and the workspace's team rules in each session.",
"author": {
"name": "AltimateAI",
Expand All @@ -25,40 +25,46 @@
"dbt_mods": {
"type": "boolean",
"title": "dbt helpers",
"description": "Compile and validate each model Claude edits, Explain and Fix on a failed dbt run, the Query results pane, model context on read, and the dbt environment doctor at session start. They run the project's own dbt.",
"description": "Compile and validate each model Claude edits, Explain and Fix on a failed dbt run, the Query results pane, model context on read, and the dbt environment doctor at session start. They run the project's own dbt. Optional; default: on.",
"default": true
},
"maxScanBytes": {
"type": "number",
"title": "Largest BigQuery scan without asking (not used yet)",
"description": "Not used yet: the BigQuery scan check isn't in this version, so no run is held. When it ships, a dbt run, build or show that BigQuery estimates would scan more bytes than this waits for your approval (0 never waits; other warehouses are never held).",
"title": "Largest BigQuery scan without asking",
"description": "A dbt run, build or show that a BigQuery dry run estimates would scan more bytes than this is held until you approve it with /allow-scan; 0 never holds, and other warehouses are never held. Optional; default: 10737418240 (10 GiB).",
"default": 10737418240,
"min": 0
},
"healthBlocksPush": {
"type": "boolean",
"title": "Project health blocks git push",
"description": "On refuses a git push while the project health check finds new issues in the models you changed this session; off only shows them.",
"description": "On refuses a git push while the project health check finds new issues in the models you changed this session, and while the check cannot run or has not finished; off only shows them. Optional; default: off.",
"default": false
},
"healthConfigPath": {
"type": "string",
"title": "Project health check config",
"description": "Path to a DataPilot health check config file, relative to the dbt project. Empty uses the default checks.",
"description": "Path to a DataPilot health check config file, relative to the dbt project. Optional; default: empty, which uses the default checks.",
"default": ""
},
"commentsPollSeconds": {
"type": "number",
"title": "Team comments refresh (seconds)",
"description": "Not used yet: team comments aren't in this version. How often the team's comments on shared docs are fetched again.",
"description": "How often, in seconds, the team's comments on shared docs are fetched again in an interactive session (30 to 86400). Optional; default: 900.",
"default": 900,
"min": 30
},
"commentsInContext": {
"type": "boolean",
"title": "Team comments in Claude's context",
"description": "Not used yet: team comments aren't in this version. When on, the open comments on a model Claude reads or edits are added to its context. Off keeps them out of Claude's context.",
"description": "When on, the open team comments on a model Claude reads or edits are added to its context; off keeps them out of Claude's context. Optional; default: on.",
"default": true
},
"show_edit_impact": {
"type": "boolean",
"title": "Show edit impact notes",
"description": "When on, the note on what a model edit reaches downstream (read by N models, dashboards) is also shown in the transcript; off sends it to Claude and the debug log only. Optional; default: off.",
"default": false
}
}
}
14 changes: 14 additions & 0 deletions plugins/altimate-code/CHANGELOG.md
Original file line number Diff line number Diff line change
Expand Up @@ -2,6 +2,20 @@

All notable changes to this plugin will be documented in this file.

## [1.8.1] — 2026-11-19

### Fixed
- The end-of-task column summary no longer says a changed column "reaches 0 columns in 0 models" when nothing was traced. When the models that read the column have no compiled SQL (a manifest from `dbt parse`), the band says "not traced: N downstream models have no compiled SQL (run `dbt compile`)". The report says the same in its downstream section, and `/lineage-report <model>.<column>` on a column the manifest does not reach is a plain error. The per-edit impact note (to Claude, and to you with `show_edit_impact`) follows the same rules. Models read in full that the value was not traced to still show as "may change" beside a reach of 0. Every changed column is asked of the manifest by name; one the SQL from before the task did not name and the manifest lacks (likely added) is said to have an unknown reach. The old name of a rename is still traced. A model read only in part counts as unread. A column found with nothing downstream still says it reaches 0.
- A missing altimate-code, or one without the `dbt_health` tool, no longer pops a toast at session start or after a turn; `/health` says it, naming the tool and the install command instead of a version.
- The settings say what each option does and that it is optional, with its default. The BigQuery scan limit and the team comments options no longer say they are unused.
- With a column in focus, the lineage pane counts only the tests on that column of the model ("3 tests on amount"); with none, or when the graph does not say which column each test checks (a manifest over 4 MiB), the tests downstream as before.
- Install and update instructions in the commands and skills give one method: `npm i -g @altimateai/altimate-code`.

### Added
- At session start, an altimate-code older than the plugin needs (or none on PATH) keeps a part at the end of the status line (a preview build, 0.0.0-…, says nothing) with its version, the binary and where it is, and `npm i -g @altimateai/altimate-code`.
- When the terminal is too narrow for the lineage pane, a line above the prompt says so, with the width it needs, until the pane can open or `/lineage` is used.
- `show_edit_impact` (off by default): also shows the per-edit impact note Claude gets in the transcript.

## [1.8.0] — 2026-11-17

### Added
Expand Down
22 changes: 12 additions & 10 deletions plugins/altimate-code/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -30,14 +30,15 @@ Then restart Claude Code.

## Prerequisites

The altimate-code CLI must be installed and configured on your PATH, version 0.12.11 or later. Older releases give no column answer, so the end-of-task summary and report can't say which columns a change reaches. Either install method works:
The altimate-code CLI must be installed and configured on your PATH, version 0.12.11 or later. Older releases give no column answer, so the end-of-task summary and report can't say which columns a change reaches. Install it with npm:

```bash
npm install -g altimate-code # requires Node 20+; exposes `altimate` and `altimate-code`
curl -fsSL https://www.altimate.sh/install | bash # standalone binary; exposes `altimate` and `altimate-code`
altimate # launch the TUI, then run /connect to configure your LLM provider
npm i -g @altimateai/altimate-code # requires Node 20+; exposes `altimate` and `altimate-code`
altimate # launch the TUI, then run /connect to configure your LLM provider
```

At session start the plugin checks the version on PATH. An older or missing altimate-code stays on the status line, with the binary, its path and this install command.

The CLI mode invokes `altimate`, the CLI's canonical name, and falls back to `altimate-code` for npm installs older than v0.7.1, so standalone installs work without an alias.

**MCP mode and the CLI name.** The MCP registration launches `altimate-code mcp serve`. npm, Homebrew and current curl and PowerShell installs expose both names, but curl and PowerShell installs older than v0.12.8 expose only `altimate`: there the server cannot start until `altimate-code` is on PATH (for example `ln -s "$(command -v altimate)" ~/.local/bin/altimate-code`), and Claude uses the CLI mode meanwhile. MCP mode also needs a build that contains `mcp serve`; check with `altimate-code mcp serve --help`. Direct MCP and `tool run` need no LLM provider; only delegation (`altimate run`) does.
Expand Down Expand Up @@ -114,7 +115,7 @@ The plugin's lineage pane follows the dbt model Claude reads or edits (by Read,
- **Warehouse lineage.** When a connected MCP server offers Altimate's `table_lineage` (Snowflake) or `databricks_table_lineage`, the pane adds what the warehouse's query history shows for the model's table. The plugin calls a tool of that name on whichever connected server offers it, itself and without a permission prompt, and sends it the model's `database.schema.table`. It does so only when you ask for the model (`/lineage`, a press in the pane, or `r`), never when the pane follows one of Claude's reads.
- **Claude's own lineage calls.** A `dbt_lineage` result is drawn as a list of edges colored by how each column is derived; ctrl+o keeps the original output under it.

`/lineage <model>[.column]` opens the pane on a model, at any width. When the pane follows Claude's reads on its own, Claude Code draws it only in a terminal at least 144 columns wide (110 once you have opened it yourself); in a narrower one it waits, and a toast says so once. Lineage calls go to the plugin's MCP server without a permission prompt, or to `altimate tool run` when the server is not available. A manifest over 4 MiB is read through altimate-code (`dbt_lineage` with `include_graph`). Run `dbt parse` or `dbt compile` first; column lineage and reach need compiled SQL. When the shown model's `.sql` file changed after the manifest was written, the pane and the edit note say so: run `dbt compile`, then `r`.
`/lineage <model>[.column]` opens the pane on a model, at any width. When the pane follows Claude's reads on its own, Claude Code draws it only in a terminal at least 144 columns wide (110 once you have opened it yourself); in a narrower one it waits, and a toast and a line above the prompt say so until the pane can open. The remedy: make the text smaller (Cmd -) or type `/lineage`. The pane's line `N tests` counts the tests on the model and on the models downstream of it; with a column selected it counts only the tests on that column of the model (`3 tests on amount`), when the graph says which column each test checks (a manifest over 4 MiB, read through altimate-code, keeps the first count). Lineage calls go to the plugin's MCP server without a permission prompt, or to `altimate tool run` when the server is not available. A manifest over 4 MiB is read through altimate-code (`dbt_lineage` with `include_graph`). Run `dbt parse` or `dbt compile` first; column lineage and reach need compiled SQL. When the shown model's `.sql` file changed after the manifest was written, the pane and the edit note say so: run `dbt compile`, then `r`.

### Team rules

Expand Down Expand Up @@ -153,16 +154,17 @@ Compile and validate runs a query on your warehouse after every model edit (`dbt

## Settings

The plugin has six settings, shown when you enable it (`userConfig` in `.claude-plugin/plugin.json`). `dbt_mods` is used by the dbt helpers, and `healthBlocksPush` and `healthConfigPath` by the project health check. `maxScanBytes` and the comments settings are not used yet: the features that read them are not in this version, so changing one changes nothing. The table says what each controls.
The plugin has seven settings, shown when you enable it (`userConfig` in `.claude-plugin/plugin.json`). All are optional: Claude Code may report them as "not set", and each then takes the default below. `dbt_mods` is used by the dbt helpers, `maxScanBytes` by the BigQuery scan check, `healthBlocksPush` and `healthConfigPath` by the project health check, the two comments settings by team comments, and `show_edit_impact` by the per-edit impact note.

| Setting | Default | What it controls |
| --- | --- | --- |
| `dbt_mods` | on | Turns all of the dbt helpers (above) off when false. |
| `maxScanBytes` | 10 GiB | The largest BigQuery scan a dbt run, build or show may make before it waits for your approval (0 never waits; at least 0). Other warehouses are never held. |
| `healthBlocksPush` | off | When on, `git push` is refused while the project health check finds new issues in the models you changed. Off, the issues are only shown. If the first check of a session lands after you changed a model, that model's existing issues count as new (the baseline then lacks them). The reverse also holds: a model you drop from a schema file before that first check lands has its new issues taken as existing, so a push is not held for them. |
| `maxScanBytes` | 10737418240 (10 GiB) | A dbt run, build or show that a BigQuery dry run estimates would scan more bytes than this is held until you approve it with `/allow-scan` (0 never holds; at least 0). Other warehouses are never held. |
| `healthBlocksPush` | off | When on, `git push` is refused while the project health check finds new issues in the models you changed, and while the check cannot run (altimate-code missing, or without `dbt_health`) or has not finished. Off, the issues are only shown. If the first check of a session lands after you changed a model, that model's existing issues count as new (the baseline then lacks them). The reverse also holds: a model you drop from a schema file before that first check lands has its new issues taken as existing, so a push is not held for them. |
| `healthConfigPath` | empty | The DataPilot health check config file, relative to the dbt project. Empty uses the default checks. |
| `commentsPollSeconds` | 900 | How often the team's comments on shared docs are fetched again (at least 30). |
| `commentsPollSeconds` | 900 | How often, in seconds, the team's comments on shared docs are fetched again in an interactive session (30 to 86400). |
| `commentsInContext` | on | Whether the open comments on a model Claude reads or edits are added to its context. |
| `show_edit_impact` | off | When on, the note on what a model edit reaches downstream is also shown in the transcript; off sends it to Claude and the debug log only. |

## Example prompts

Expand All @@ -187,7 +189,7 @@ Full details: **[PRIVACY.md](../../PRIVACY.md)** · Security policy and disclosu

| Symptom | Fix |
|---|---|
| Neither `altimate-code` nor `altimate` on PATH | `npm install -g altimate-code` (Node 20+) or `curl -fsSL https://www.altimate.sh/install \| bash`, then run `altimate` once to configure your provider. |
| Neither `altimate-code` nor `altimate` on PATH | `npm i -g @altimateai/altimate-code` (Node 20+), then run `altimate` once to configure your provider. |
| MCP tools missing from the session | Managed settings may disallow MCP servers, `altimate-code` may not be on PATH (curl and PowerShell installs older than v0.12.8 expose only `altimate`), or the build lacks `mcp serve`. Claude falls back to `altimate tool run`, so no action is required. |
| `Unauthorized` / `No provider configured` | Run `altimate` (`altimate-code` on npm installs older than v0.7.1) to open the TUI and reconfigure your provider via `/connect`, or set the provider API key env var. |
| Task hangs for several minutes | Open `altimate` (`altimate-code` on npm installs older than v0.7.1) to check for a pending prompt in the TUI, or re-run specifying a known-good model. |
Expand Down
2 changes: 1 addition & 1 deletion plugins/altimate-code/commands/altimate.md
Original file line number Diff line number Diff line change
Expand Up @@ -14,7 +14,7 @@ For delegation you MUST run the task via the `altimate-code` CLI — do NOT atte
Delegation workflow (follow in order, no skipping):

1. **Verify altimate-code is installed.** There is no separate check: the run command in step 3 resolves the CLI first, ignoring aliases and shell functions, and exits 127 with `altimate-code CLI not found` before doing anything else. The CLI ships under two names: npm and Homebrew installs expose both `altimate` and `altimate-code`, while older curl and PowerShell installs expose only `altimate`. Either one is fine. If the run exits with `altimate-code CLI not found`, stop and tell the user:
> altimate-code is not installed. Install with `npm install -g altimate-code` (Node 20+) or `curl -fsSL https://www.altimate.sh/install | bash`, then run `altimate` once to configure your provider/warehouse auth, then re-run `/altimate-code:altimate <task>`.
> altimate-code is not installed. Install with `npm i -g @altimateai/altimate-code` (Node 20+), then run `altimate` once to configure your provider/warehouse auth, then re-run `/altimate-code:altimate <task>`.

2. **Vague debug** ("X is broken") with no failure named: **stop — ask the user for the specific error before delegating.**

Expand Down
4 changes: 2 additions & 2 deletions plugins/altimate-code/commands/review.md
Original file line number Diff line number Diff line change
Expand Up @@ -17,7 +17,7 @@ Run the `altimate-code review` CLI on the current dbt project and present the si

## Workflow

1. **Verify altimate-code is on PATH.** There is no separate check: the resolver below runs first in the same Bash call as the review, before any `gh pr checkout`. If it exits 127 with `altimate-code CLI not found`, tell the user: `"altimate-code is not installed. Install with 'npm install -g altimate-code' (Node 20+) or 'curl -fsSL https://www.altimate.sh/install | bash', then run 'altimate' once to configure your provider and warehouse auth."` — then STOP.
1. **Verify altimate-code is on PATH.** There is no separate check: the resolver below runs first in the same Bash call as the review, before any `gh pr checkout`. If it exits 127 with `altimate-code CLI not found`, tell the user: `"altimate-code is not installed. Install with 'npm i -g @altimateai/altimate-code' (Node 20+), then run 'altimate' once to configure your provider and warehouse auth."` — then STOP.

The CLI ships under two names: npm and Homebrew installs expose both `altimate` and `altimate-code`, while older curl and PowerShell installs expose only `altimate`. Every review command below runs through `"$ALTIMATE_BIN"`, in the same Bash call as this resolver (shell state does not persist between calls). The lookup ignores aliases and shell functions, so a user alias such as `alias altimate-code=altimate` cannot be captured as the binary path:

Expand Down Expand Up @@ -65,7 +65,7 @@ Anything else (`--manifest`, `--cwd`) only if the user explicitly requested it o

| Symptom | Message to user |
|---|---|
| `altimate-code CLI not found` (neither `altimate-code` nor `altimate` on PATH) | "altimate-code is not installed. Install with `npm install -g altimate-code` (Node 20+) or `curl -fsSL https://www.altimate.sh/install \| bash`, then run `altimate` once to configure auth." |
| `altimate-code CLI not found` (neither `altimate-code` nor `altimate` on PATH) | "altimate-code is not installed. Install with `npm i -g @altimateai/altimate-code` (Node 20+), then run `altimate` once to configure auth." |
| `Unauthorized: Incorrect auth token` | "altimate-code's LLM provider auth is misconfigured. Run `altimate` (`altimate-code` on npm installs older than v0.7.1) to reconfigure, or re-run with `--no-ai`." |
| `fatal: not a git repository` | "This directory isn't a git repo. `altimate-code review` needs git to compute the diff." |
| No `dbt_project.yml` | "No dbt project detected at this path. Run from the directory with `dbt_project.yml`." |
Expand Down
Loading
Loading