Lampion MCP Server
Serveur MCP (Model Context Protocol) pour Lampion — permet aux agents IA (Claude, Cursor, Windsurf, etc.) de gerer des bases de donnees PostgreSQL managees.
Installation
pip install lampion-mcp
Depuis les sources (développement)
git clone https://github.com/Sylv-Lej/lampion-mcp.git
cd lampion-mcp
pip install -e '.[dev]'
pytest tests/
Usage
# Via CLI
lampion-mcp --token lmp_live_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
# Via env vars
export LAMPION_TOKEN=lmp_live_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
export LAMPION_API_URL=https://api.lampion.cloud # optionnel
lampion-mcp
Configuration Claude Desktop
Fichier ~/.claude/claude_desktop_config.json :
{
"mcpServers": {
"lampion": {
"command": "lampion-mcp",
"env": {
"LAMPION_TOKEN": "lmp_live_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
}
}
}
}
Configuration Cursor / VS Code
Fichier .cursor/mcp.json ou .vscode/mcp.json :
{
"servers": {
"lampion": {
"command": "lampion-mcp",
"env": {
"LAMPION_TOKEN": "lmp_live_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
}
}
}
}
Tools disponibles (47)
Projets
list_projects— Lister les projetsget_project— Details d'un projetcreate_project— Creer un projet PostgreSQL (choix de region : Scaleway Paris ou OVH Gravelines)delete_project— Supprimer un projet (admin requis)
Branches
list_branches— Lister les branchescreate_branch— Creer une branche copy-on-writedelete_branch— Supprimer une branchereset_branch— Reset une branche a l'etat du parent (comme git reset --hard)merge_branch— Merger les changements de schema vers la branche principale (dry_run possible)promote_branch— Promouvoir une branche en nouvelle branche principale (destructif)
Computes
list_endpoints— Lister les endpoints avec connection stringsget_endpoint— Details d'un endpointsuspend_endpoint— Suspendre (scale to zero)resume_endpoint— Relancer un compute
SQL
execute_sql— Executer une requete SQLget_schema— Introspecter le schema (databases -> schemas -> tables -> colonnes)
Databases & Roles
list_databases/create_database/drop_databaselist_roles/create_role/drop_role
Extensions
list_extensions/install_extension/uninstall_extension
Observabilite
get_metrics— Metriques temps reel (CPU, RAM, connections, cache)get_slow_queries— Top queries lentes (pg_stat_statements)get_logs— Tail des logs PostgreSQLget_project_usage/get_org_usage— Usage et couts
Backups
list_backups/create_backup/restore_backup
Dump
pg_dump— Export SQL d'une database
Anonymisation (PII Masking)
get_anonymization— Statut et regles de masquage d'une brancheenable_anonymization— Activer le masquage dynamique (cree un role lampion_analyst)disable_anonymization— Desactiver le masquagelist_masking_functions— Fonctions disponibles (fake_email, hash, partial, etc.)list_masking_rules— Regles actives (schema.table.column -> fonction)add_masking_rule— Ajouter une regle de masquagedelete_masking_rule— Supprimer une regle
Securite
get_allowlist— IP allowlist d'un endpointset_allowlist— Configurer l'IP allowlistlist_permissions— Permissions table-level par rolegrant_permission— GRANT un privilege (SELECT, INSERT, UPDATE, DELETE, ALL)revoke_permission— REVOKE un privilege
Compliance
get_residency— Ou sont les donnees physiquement (datacenter, provider, chiffrement, certifications)
Resources MCP (5)
lampion://projects— Liste des projetslampion://projects/{id}— Details d'un projetlampion://projects/{id}/endpoints— Endpoints avec connection stringslampion://projects/{id}/branches— Branches d'un projetlampion://usage— Usage organisation et estimation des couts
Exemples d'utilisation par un agent
"Cree un projet PostgreSQL pour mon app de blog"
L'agent appellera create_project(name="blog-db") puis list_endpoints() pour obtenir la connection string.
"Montre-moi le schema de ma base de donnees"
L'agent appellera get_schema(project_id="xxx", database="postgres", schema="public").
"Quelle est ma requete la plus lente ?"
L'agent appellera get_slow_queries(project_id="xxx", endpoint_id="ep-xxx").
"Anonymise les emails de la table users"
L'agent appellera enable_anonymization() puis add_masking_rule(table_name="users", column_name="email", masking_function="anon.fake_email()").
"Merge les changements de schema de ma branche dev vers main"
L'agent appellera merge_branch(project_id="xxx", branch_id="dev-id", dry_run=True) pour previsualiser, puis merge_branch(..., dry_run=False) pour executer.
Generation d'API Key
- Se connecter sur https://app.lampion.cloud
- Aller dans Settings -> API Keys
- Creer une cle (prefixe
lmp_live_) - Copier la cle et la configurer dans le MCP server
Release files for lampion-mcp 0.1.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| lampion_mcp-0.1.0.tar.gz | 13.1 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| lampion_mcp-0.1.0-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 24.6 kB
Release files / lampion_mcp-0.1.0.tar.gz
| Download URL | lampion_mcp-0.1.0.tar.gz |
|---|---|
| Size | 13.1 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
1626e90e636120ee4488ceab1e61782bf61a65ab336cf271ad375daf07df266b
|
|
BLAKE2b-256 checksum How to use checksums |
a703428a31b2d024caa0524b42fb9f04bccfcf2921bd1414629296e414983492
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 20, 2026.
Transparency logRelease files / lampion_mcp-0.1.0-py3-none-any.whl
| Download URL | lampion_mcp-0.1.0-py3-none-any.whl |
|---|---|
| Size | 11.5 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
09c8033fbcff2291595c31e14c4d9721b4d716e5e5e6564ab12504d616a0f0e7
|
|
BLAKE2b-256 checksum How to use checksums |
8cc602b796d58e199f26bd480dafd578b426dccd6e696f3ac3306186b522bd0b
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 20, 2026.
Transparency log