AnomalyArmor MCP Server - Data observability tools for AI assistants
Project description
armor-mcp
The AnomalyArmor MCP server lets AI assistants (Claude Code, Cursor, Claude Desktop, any MCP client) interact with your data warehouse through 50+ structured tools: query alerts, monitor freshness, investigate schema drift, configure validity rules, and trace lineage in natural language.
- Homepage: https://www.anomalyarmor.ai
- Docs: https://docs.anomalyarmor.ai/integrations/mcp-server
- Server card: https://www.anomalyarmor.ai/.well-known/mcp/server-card.json
- Status: Production. Versioned MCP protocol
2025-06-18. OAuth 2.1 + API key auth.
Two ways to connect
| Method | Auth | Best for |
|---|---|---|
| Remote (recommended) | OAuth 2.1 (Clerk) | Zero install, always up to date |
| Local | API key | Air-gapped, on-call laptops, custom proxies |
Remote (recommended)
The server is hosted at https://mcp.anomalyarmor.ai/mcp (Streamable HTTP transport). First call opens a browser for OAuth; the token is cached for ~12h.
Claude Code
claude mcp add anomalyarmor --transport http https://mcp.anomalyarmor.ai/mcp
Cursor
Add to ~/Library/Application Support/Cursor/mcp.json (macOS), ~/.config/Cursor/mcp.json (Linux), or %APPDATA%\Cursor\mcp.json (Windows):
{
"mcpServers": {
"anomalyarmor": {
"url": "https://mcp.anomalyarmor.ai/mcp"
}
}
}
Any MCP client
URL: https://mcp.anomalyarmor.ai/mcp
Transport: streamable-http
Auth: OAuth 2.1
Discovery: https://mcp.anomalyarmor.ai/.well-known/oauth-protected-resource
Local (API key)
Use when the remote server isn't reachable or when you want no third-party dependency.
Install
# Recommended: uvx (no install)
uvx armor-mcp
# Or pip
pip install armor-mcp
Configure
Get an API key from Settings → API Keys at https://app.anomalyarmor.ai.
export ARMOR_API_KEY=aa_live_your_key_here
Or persist to ~/.armor/config.yaml:
api_key: aa_live_your_key_here
Wire to a client
{
"mcpServers": {
"anomalyarmor": {
"command": "uvx",
"args": ["armor-mcp"],
"env": {
"ARMOR_API_KEY": "aa_live_your_key_here"
}
}
}
}
Tools
53 tools across 8 categories. Every tool ships with MCP ToolAnnotations (readOnlyHint, destructiveHint, idempotentHint, openWorldHint: false) so hosts can decide when to auto-confirm.
Health & Briefings
| Tool | Description |
|---|---|
health_summary |
Overall warehouse health: stale assets, drift, alerts |
get_todays_briefing |
Daily digest with key insights |
get_coverage |
Monitoring coverage analysis |
manage_coverage |
Update coverage targets |
recommend |
AI-suggested next monitoring actions |
Alerts & Incidents
| Tool | Description |
|---|---|
list_alerts |
Query alerts with filters (severity, date, status) |
list_inbox_alerts |
Alerts not yet triaged |
get_alerts_summary |
Aggregate counts and trends |
get_alert_trends |
Period-over-period trend analysis |
get_alert_history |
Activity timeline for an alert |
update_alert |
Acknowledge / resolve / dismiss / snooze |
list_alert_rules |
View configured rules |
create_alert_rule |
Add a new alert rule |
manage_alert_rule |
Update / delete / preview rules |
manage_rule_destinations |
Wire rules to destinations |
Assets
| Tool | Description |
|---|---|
list_assets |
List data sources with filters |
create_asset |
Connect a new source |
manage_asset |
Update / delete / test connection |
trigger_asset_discovery |
Start schema discovery (async) |
Freshness
| Tool | Description |
|---|---|
get_freshness_summary |
Freshness overview |
check_freshness |
Check a specific asset / table |
setup_freshness |
Bulk-create freshness schedules |
list_freshness_schedules |
View configured schedules |
manage_freshness_schedule |
Update / delete a schedule |
Schema Monitoring
| Tool | Description |
|---|---|
get_schema_summary |
Drift overview |
list_schema_changes |
Recent changes with severity |
create_schema_baseline |
Capture current schema as a baseline |
enable_schema_monitoring |
Start drift detection |
disable_schema_monitoring |
Stop drift detection |
Data Quality (Metrics & Validity)
| Tool | Description |
|---|---|
get_metrics_summary |
Metric health by asset |
list_metrics |
List configured metrics |
create_metric |
Add a metric (row count, null %, etc.) |
manage_metric |
Update / delete / capture a metric |
get_validity_summary |
Pass/fail rate for validity rules |
list_validity_rules |
List validity rules |
create_validity_rule |
Add a NOT_NULL / UNIQUE / RANGE / REGEX rule |
manage_validity_rule |
Update / delete / run a rule |
create_referential_check |
Cross-table referential check |
manage_referential |
Update / delete a referential check |
Destinations & Routing
| Tool | Description |
|---|---|
list_destinations |
Slack / email / webhook / PagerDuty configs |
setup_destination |
Add a new destination |
manage_destination |
Update / delete / test |
Intelligence, Lineage, Tags, Jobs
| Tool | Description |
|---|---|
ask_question |
Synchronous natural-language Q&A about your data |
generate_intelligence |
Trigger AI analysis (async, expensive) |
get_lineage |
Upstream/downstream dependency graph |
list_tags |
Tags for an asset |
create_tag |
Tag a table or column |
apply_tags |
Bulk tag application |
job_status |
Track an async job |
cancel_job |
Cancel a running async job |
get_api_key_info |
Inspect the current API key's scope |
Full descriptions, parameter schemas, and example prompts: https://docs.anomalyarmor.ai/integrations/mcp-server.
Tool safety conventions
Hosts respect the annotations:
| Category | Pattern | Confirmation |
|---|---|---|
| Read-only | list_*, get_*, check_*, search_* |
Auto-invoke |
| Mutating | create_*, update_*, enable_*, disable_*, manage_* |
Confirm with user |
| Destructive | delete_* (via manage_*), dismiss_alert |
Always confirm; list affected entities |
| Expensive | generate_intelligence, trigger_asset_discovery |
Confirm; these queue paid jobs |
Multi-tenancy
The OAuth token (or API key) is scoped to the user's company. You cannot impersonate another tenant. Don't fabricate company_id arguments.
Async jobs
trigger_asset_discovery, generate_intelligence, and metric captures return a job_id. Poll with job_status(job_id) rather than re-triggering.
Discovery surfaces
| File | Purpose |
|---|---|
https://mcp.anomalyarmor.ai/.well-known/oauth-protected-resource |
RFC 9728 protected-resource metadata |
https://www.anomalyarmor.ai/.well-known/mcp/server-card.json |
SEP-1649 MCP server card |
https://www.anomalyarmor.ai/.well-known/agent-card.json |
A2A v0.2 agent card |
https://www.anomalyarmor.ai/.well-known/api-catalog |
RFC 9727 API catalog linkset |
Development
git clone https://github.com/anomalyarmor/agents.git
cd agents/armor-mcp
uv sync
uv run pytest
License
MIT — see LICENSE.
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file armor_mcp-0.6.1.tar.gz.
File metadata
- Download URL: armor_mcp-0.6.1.tar.gz
- Upload date:
- Size: 31.0 kB
- Tags: Source
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/6.1.0 CPython/3.13.12
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
25d1c940dbd5cf6af188b337540c7ee8b2492ac6abd4571a994b24e8c81573ab
|
|
| MD5 |
d1d344d3606d71e313275041dacf8fdf
|
|
| BLAKE2b-256 |
4389438ceb7dd8f140b1b91848e97e6bbf6115faed3ec8f0983772a9a66d9c63
|
Provenance
The following attestation bundles were made for armor_mcp-0.6.1.tar.gz:
Publisher:
publish-pypi.yml on anomalyarmor/agents
-
Statement:
-
Statement type:
https://in-toto.io/Statement/v1 -
Predicate type:
https://docs.pypi.org/attestations/publish/v1 -
Subject name:
armor_mcp-0.6.1.tar.gz -
Subject digest:
25d1c940dbd5cf6af188b337540c7ee8b2492ac6abd4571a994b24e8c81573ab - Sigstore transparency entry: 1333844257
- Sigstore integration time:
-
Permalink:
anomalyarmor/agents@352e8dbc2c49aac1401edacd3d24758dc2b1f752 -
Branch / Tag:
refs/tags/v0.6.1 - Owner: https://github.com/anomalyarmor
-
Access:
public
-
Token Issuer:
https://token.actions.githubusercontent.com -
Runner Environment:
github-hosted -
Publication workflow:
publish-pypi.yml@352e8dbc2c49aac1401edacd3d24758dc2b1f752 -
Trigger Event:
release
-
Statement type:
File details
Details for the file armor_mcp-0.6.1-py3-none-any.whl.
File metadata
- Download URL: armor_mcp-0.6.1-py3-none-any.whl
- Upload date:
- Size: 33.1 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/6.1.0 CPython/3.13.12
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
81ce4ba87523e9e485a7149b8338e1388c7cf7e2712d1e0628bea25ae1426828
|
|
| MD5 |
56ee7cabd10a0d311f2ed2330159c7ad
|
|
| BLAKE2b-256 |
b1b1f04ae4ec682b8b2b4651b4fc9ba6c3f547802f4e0d19842aa2d72ef39636
|
Provenance
The following attestation bundles were made for armor_mcp-0.6.1-py3-none-any.whl:
Publisher:
publish-pypi.yml on anomalyarmor/agents
-
Statement:
-
Statement type:
https://in-toto.io/Statement/v1 -
Predicate type:
https://docs.pypi.org/attestations/publish/v1 -
Subject name:
armor_mcp-0.6.1-py3-none-any.whl -
Subject digest:
81ce4ba87523e9e485a7149b8338e1388c7cf7e2712d1e0628bea25ae1426828 - Sigstore transparency entry: 1333844330
- Sigstore integration time:
-
Permalink:
anomalyarmor/agents@352e8dbc2c49aac1401edacd3d24758dc2b1f752 -
Branch / Tag:
refs/tags/v0.6.1 - Owner: https://github.com/anomalyarmor
-
Access:
public
-
Token Issuer:
https://token.actions.githubusercontent.com -
Runner Environment:
github-hosted -
Publication workflow:
publish-pypi.yml@352e8dbc2c49aac1401edacd3d24758dc2b1f752 -
Trigger Event:
release
-
Statement type: