vibedasher-mcp 0.1.0__tar.gz

vibedasher_mcp-0.1.0/.gitignore

@@ -0,0 +1,143 @@
+# Node (Nuxt.js)
+node_modules/
+
+# Side-lambda build artifacts — deploy_side_lambdas() builds in a temp dir
+# (mktemp), but legacy in-place runs left these behind; never commit them.
+lambdas/*/package-lock.json
+lambdas/*/deployment.zip
+
+# Claude Code local backups (contains secrets — never commit)
+.claude/backups/
+test-results/
+frontend/playwright-report
+scheduled_tasks.lock
+playwright-report
+.playwright-mcp/
+.npm/
+yarn.lock
+pnpm-lock.yaml
+dist/
+.nuxt/
+output/
+.env
+.env.*
+.DS_Store
+aider
+api/build/
+.lambda-builds/
+api/debug_logs_json/
+api/seeds/shadcn-source/
+bi_test_datasets/
+build/cron/
+api/debug_logs/
+
+# Python (FastAPI)
+__pycache__/
+*.py[cod]
+*.pyo
+*.pyd
+*.pyc
+*.pyo
+*.pyd
+.Python
+env/
+venv/
+ENV/
+.venv/
+*.egg-info/
+.eggs/
+*.egg
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Tauri
+/src-tauri/target/
+/src-tauri/.cargo/
+/src-tauri/Cargo.lock
+/src-tauri/Cargo.toml.orig
+/src-tauri/.tauri-build
+/src-tauri/.tauri
+
+# VSCode
+.vscode/
+
+# Terraform
+.terraform/
+*.tfstate
+*.tfstate.*
+crash.log
+# Allow environment-specific tfvars in infra/
+!infra/*.tfvars
+!infra/**/*.tfvars
+override.tf
+override.tf.json
+*_override.tf
+*_override.tf.json
+.terraformrc
+terraform.rc
+
+# OS
+Thumbs.db
+.DS_Store
+
+# Logs
+*.log
+
+# Misc
+.idea/
+*.swp
+*.swo
+
+# Local DBs
+*.db
+
+# Test/coverage
+coverage/
+*.cover
+*.pyc
+*.pyo
+*.pyd
+.cache/
+
+# Lambda/Serverless
+.serverless/
+
+# Docker
+*.pid
+*.pid.lock
+
+# Others
+*.bak
+*.tmp
+
+# Ignore local secrets
+secrets.*
+api/*.local.env
+
+# SSL certificates (never commit private keys!)
+nginx/ssl/*.pem
+nginx/ssl/*.key
+nginx/ssl/*.crt
+
+# Production build outputs
+dist/
+
+# On-premise configuration (contains secrets)
+.env
+config/*.env
+!config/*.env.example
+
+.agents
+skills
+_bmad
+.envrc
+.vercel
+HARDENING/screenshots/s3b-session.har
+HARDENING/screenshots/*.har
+
+# i18n_sync.py dispatch manifest — regenerated on demand, not source
+i18n-delta.json
+
+# `skills` above swallows api/seeds/skills/** — those are lambda-bundled seed assets and MUST be tracked
+!api/seeds/skills/
+!api/seeds/skills/**

vibedasher_mcp-0.1.0/PKG-INFO

@@ -0,0 +1,99 @@
+Metadata-Version: 2.4
+Name: vibedasher-mcp
+Version: 0.1.0
+Summary: MCP server exposing the Vibedasher data engine (datasets, query, viz) to a customer's AI.
+Project-URL: Homepage, https://vibedasher.com
+Project-URL: Source, https://github.com/JulienGdnr/vibedasher
+Author: Vibedasher
+License: MIT
+Keywords: analytics,bi,eject,mcp,model-context-protocol,vibedasher
+Requires-Python: >=3.10
+Requires-Dist: fastmcp>=2.0.0
+Requires-Dist: httpx<0.29.0,>=0.23.0
+Requires-Dist: msgpack>=1.0.0
+Requires-Dist: vibedasher>=2.0.0
+Provides-Extra: dev
+Requires-Dist: pytest>=7.0; extra == 'dev'
+Description-Content-Type: text/markdown
+
+# vibedasher-mcp
+
+An [MCP](https://modelcontextprotocol.io) server that exposes the **Vibedasher data
+engine** — datasets, CSV upload + ETL, headless SQL query, and dashboard (viz)
+management — as tools a customer's AI (Claude Code, Cursor, ...) can drive directly.
+
+This is the **control plane** (VD-601, PIVOT/PLAN.md EP-6). It wraps the published
+[`vibedasher` Python SDK](https://pypi.org/project/vibedasher/) and calls only the public, metered `/api/v1/*`
+API. The *eject* tools (pull a viz's code + SDK-wiring manifest so your AI recreates
+the dashboard natively in your stack) ship separately as VD-602.
+
+## Tools
+
+| Tool | What it does |
+|------|--------------|
+| `list_datasets` | List datasets the key can read (id, name, `cleanSQLName`, status). |
+| `get_dataset(dataset_id)` | One dataset's metadata + column schema. |
+| `upload_dataset(name, csv_content, ...)` | Create a CSV dataset, upload, load, poll to READY. |
+| `run_query(dataset_ids, sql, params)` | Inline alias-only SQL across your datasets (plural `datasetIds`, joins allowed) → typed columns + rows. |
+| `list_vizzes(include_unpublished=False)` | List dashboards. |
+| `get_viz(viz_id)` | One viz's metadata. |
+| `create_viz(name, dataset_ids, seed_id/dashboard_config, ...)` | Create a viz (AI dashboard-build entry). |
+| `build_viz(viz_id, branch)` | Compile a viz branch into a renderable bundle. |
+
+`run_query` reuses the SDK's hand-written `query()` transport (VD-301): the caller
+never sees inline-vs-presigned delivery, MessagePack, or retries — one call in,
+typed rows out. Each dataset id resolves server-side to that dataset's
+`cleanSQLName` alias under the VD-203 RLS/alias-rewrite; SQL references only aliases.
+
+## Auth
+
+API key only, via `X-Api-Key` (handled by the SDK's `create_client`). Set:
+
+```bash
+export VIBEDASHER_API_KEY=...           # mint via `POST /v1/api-keys` or the console
+export VIBEDASHER_REGION=eu-central-1   # or us-east-1
+# export VIBEDASHER_BASE_URL=...        # optional override (on-prem/staging)
+```
+
+The key is read once at first tool call and is never logged or echoed in any
+tool result.
+
+## Metering
+
+Every tool hits the public, metered endpoints (`create_dataset` is credit-gated;
+`run_query` is metered via the usage/credits event). The MCP adds no side channel
+and bypasses no metering.
+
+## Run
+
+```bash
+pip install vibedasher-mcp        # (monorepo dev: also make `vibedasher` importable)
+vibedasher-mcp                    # stdio MCP server
+# or:  python -m vibedasher_mcp
+```
+
+MCP client config (Claude Code / Cursor):
+
+```json
+{
+  "mcpServers": {
+    "vibedasher": {
+      "command": "vibedasher-mcp",
+      "env": { "VIBEDASHER_API_KEY": "...", "VIBEDASHER_REGION": "eu-central-1" }
+    }
+  }
+}
+```
+
+## Development
+
+The package depends on the sibling SDK at `../sdk/py`. Tests wire that path
+automatically (`tests/conftest.py`), so from `packages/mcp`:
+
+```bash
+python -m pytest tests/ -q
+```
+
+> Note: `fastmcp` is a **client-side** dependency of this standalone package; it is
+> not bundled into the API Lambdas, so the `export_requirements.py` step in the root
+> `CLAUDE.md` does not apply here.

vibedasher_mcp-0.1.0/README.md

@@ -0,0 +1,81 @@
+# vibedasher-mcp
+
+An [MCP](https://modelcontextprotocol.io) server that exposes the **Vibedasher data
+engine** — datasets, CSV upload + ETL, headless SQL query, and dashboard (viz)
+management — as tools a customer's AI (Claude Code, Cursor, ...) can drive directly.
+
+This is the **control plane** (VD-601, PIVOT/PLAN.md EP-6). It wraps the published
+[`vibedasher` Python SDK](https://pypi.org/project/vibedasher/) and calls only the public, metered `/api/v1/*`
+API. The *eject* tools (pull a viz's code + SDK-wiring manifest so your AI recreates
+the dashboard natively in your stack) ship separately as VD-602.
+
+## Tools
+
+| Tool | What it does |
+|------|--------------|
+| `list_datasets` | List datasets the key can read (id, name, `cleanSQLName`, status). |
+| `get_dataset(dataset_id)` | One dataset's metadata + column schema. |
+| `upload_dataset(name, csv_content, ...)` | Create a CSV dataset, upload, load, poll to READY. |
+| `run_query(dataset_ids, sql, params)` | Inline alias-only SQL across your datasets (plural `datasetIds`, joins allowed) → typed columns + rows. |
+| `list_vizzes(include_unpublished=False)` | List dashboards. |
+| `get_viz(viz_id)` | One viz's metadata. |
+| `create_viz(name, dataset_ids, seed_id/dashboard_config, ...)` | Create a viz (AI dashboard-build entry). |
+| `build_viz(viz_id, branch)` | Compile a viz branch into a renderable bundle. |
+
+`run_query` reuses the SDK's hand-written `query()` transport (VD-301): the caller
+never sees inline-vs-presigned delivery, MessagePack, or retries — one call in,
+typed rows out. Each dataset id resolves server-side to that dataset's
+`cleanSQLName` alias under the VD-203 RLS/alias-rewrite; SQL references only aliases.
+
+## Auth
+
+API key only, via `X-Api-Key` (handled by the SDK's `create_client`). Set:
+
+```bash
+export VIBEDASHER_API_KEY=...           # mint via `POST /v1/api-keys` or the console
+export VIBEDASHER_REGION=eu-central-1   # or us-east-1
+# export VIBEDASHER_BASE_URL=...        # optional override (on-prem/staging)
+```
+
+The key is read once at first tool call and is never logged or echoed in any
+tool result.
+
+## Metering
+
+Every tool hits the public, metered endpoints (`create_dataset` is credit-gated;
+`run_query` is metered via the usage/credits event). The MCP adds no side channel
+and bypasses no metering.
+
+## Run
+
+```bash
+pip install vibedasher-mcp        # (monorepo dev: also make `vibedasher` importable)
+vibedasher-mcp                    # stdio MCP server
+# or:  python -m vibedasher_mcp
+```
+
+MCP client config (Claude Code / Cursor):
+
+```json
+{
+  "mcpServers": {
+    "vibedasher": {
+      "command": "vibedasher-mcp",
+      "env": { "VIBEDASHER_API_KEY": "...", "VIBEDASHER_REGION": "eu-central-1" }
+    }
+  }
+}
+```
+
+## Development
+
+The package depends on the sibling SDK at `../sdk/py`. Tests wire that path
+automatically (`tests/conftest.py`), so from `packages/mcp`:
+
+```bash
+python -m pytest tests/ -q
+```
+
+> Note: `fastmcp` is a **client-side** dependency of this standalone package; it is
+> not bundled into the API Lambdas, so the `export_requirements.py` step in the root
+> `CLAUDE.md` does not apply here.

vibedasher_mcp-0.1.0/pyproject.toml

@@ -0,0 +1,35 @@
+[build-system]
+requires = ["hatchling"]
+build-backend = "hatchling.build"
+
+[project]
+name = "vibedasher-mcp"
+version = "0.1.0"
+description = "MCP server exposing the Vibedasher data engine (datasets, query, viz) to a customer's AI."
+readme = "README.md"
+requires-python = ">=3.10"
+license = { text = "MIT" }
+authors = [{ name = "Vibedasher" }]
+keywords = ["vibedasher", "mcp", "model-context-protocol", "bi", "analytics", "eject"]
+dependencies = [
+    "fastmcp>=2.0.0",
+    "httpx>=0.23.0,<0.29.0",
+    "msgpack>=1.0.0",
+    # The published Vibedasher SDK — the MCP wraps it for auth + the query()
+    # transport. Pinned to the current major; in this monorepo it resolves from
+    # ../sdk/py during development (see README).
+    "vibedasher>=2.0.0",
+]
+
+[project.optional-dependencies]
+dev = ["pytest>=7.0"]
+
+[project.scripts]
+vibedasher-mcp = "vibedasher_mcp.server:main"
+
+[project.urls]
+Homepage = "https://vibedasher.com"
+Source = "https://github.com/JulienGdnr/vibedasher"
+
+[tool.hatch.build.targets.wheel]
+packages = ["vibedasher_mcp"]

vibedasher_mcp-0.1.0/run-local.sh

@@ -0,0 +1,40 @@
+#!/usr/bin/env bash
+# Run the vibedasher MCP server straight from this monorepo.
+#
+# WHY THIS EXISTS: `vibedasher-mcp` is not published to PyPI yet (see
+# GOLIVE_QA/PHASE208/EJECT-QA-FINDINGS.md F-01), so the README's
+# `pip install vibedasher-mcp` cannot work and there is no `vibedasher-mcp`
+# binary on PATH. This script is the monorepo stand-in: it puts the sibling SDK
+# on PYTHONPATH and launches the stdio server.
+#
+# Once the packages ARE published, delete this and use the README's config:
+#   { "command": "vibedasher-mcp", "env": { "VIBEDASHER_API_KEY": "..." } }
+#
+# SECRETS: the API key is NEVER stored in the repo. It lives in
+# ~/.vibedasher-mcp.env (chmod 600), which this script sources. Create it with:
+#   VIBEDASHER_API_KEY=<key from POST /v1/api-keys>
+#   VIBEDASHER_REGION=eu-central-1
+
+set -euo pipefail
+
+REPO_ROOT="$(cd "$(dirname "${BASH_SOURCE[0]}")/../.." && pwd)"
+ENV_FILE="${VIBEDASHER_MCP_ENV_FILE:-$HOME/.vibedasher-mcp.env}"
+
+if [ -f "$ENV_FILE" ]; then
+  set -a
+  # shellcheck disable=SC1090
+  . "$ENV_FILE"
+  set +a
+fi
+
+if [ -z "${VIBEDASHER_API_KEY:-}" ]; then
+  echo "vibedasher-mcp: VIBEDASHER_API_KEY is not set." >&2
+  echo "  Expected it in $ENV_FILE (chmod 600) or the environment." >&2
+  echo "  Mint one with: POST /v1/api-keys (session JWT auth)." >&2
+  exit 1
+fi
+
+export VIBEDASHER_REGION="${VIBEDASHER_REGION:-eu-central-1}"
+export PYTHONPATH="$REPO_ROOT/packages/sdk/py:$REPO_ROOT/packages/mcp${PYTHONPATH:+:$PYTHONPATH}"
+
+exec python3 -m vibedasher_mcp "$@"

vibedasher_mcp-0.1.0/tests/conftest.py

@@ -0,0 +1,16 @@
+"""Dev-time path wiring so tests import the in-repo package + sibling SDK.
+
+In a real install, ``vibedasher_mcp`` and ``vibedasher`` are pip-installed and
+this file is unnecessary; it only makes the monorepo checkout runnable without an
+editable install.
+"""
+import pathlib
+import sys
+
+_MCP_ROOT = pathlib.Path(__file__).resolve().parents[1]  # packages/mcp
+_SDK_PY = _MCP_ROOT.parent / "sdk" / "py"  # packages/sdk/py
+
+for path in (_MCP_ROOT, _SDK_PY):
+    p = str(path)
+    if p not in sys.path:
+        sys.path.insert(0, p)