Connect Archer to Claude: Sync data feeds and manage system metadata
Give Claude secure, authenticated access to Archer's REST and Content APIs. Learn how to configure an MCP server, query records, and trigger workflow actions.
If you need to connect Archer to Claude to automate GRC compliance reporting, trigger data feeds, or audit security events, you need a Model Context Protocol (MCP) server. This server acts as the translation layer between Claude's tool calls and Archer's dual-headed REST and Content APIs. You can either build and maintain this integration infrastructure yourself, or use a managed integration platform like Truto to dynamically generate a secure, authenticated MCP server URL. If your team uses ChatGPT, check out our guide on connecting Archer to ChatGPT or explore our broader architectural overview on connecting Archer to AI Agents.
Giving a Large Language Model (LLM) read and write access to a highly configurable governance, risk, and compliance platform like Archer is a brutal engineering challenge. You have to handle session-based authentication tokens, parse wildly nested JSON structures for record manipulation, and manage complex cross-reference mappings. Every time an admin updates an Archer layout or adds a required field, your custom MCP server risks breaking.
This guide breaks down exactly how to use Truto to generate a secure, managed MCP server for Archer, connect it natively to Claude Desktop, and execute complex governance workflows using natural language.
The Engineering Reality of the Archer API
A custom MCP server is a self-hosted integration layer. While the open MCP standard provides a predictable way for models to discover tools, the reality of implementing it against Archer's APIs is painful. You are not just integrating a simple CRUD app - you are integrating a highly customized enterprise data engine that utilizes two entirely different API paradigms simultaneously.
If you decide to build a custom MCP server for Archer, you own the entire API lifecycle. Here are the specific challenges you will face:
Fragmented API Paradigms
Archer splits its functionality across a legacy Content API and a newer RESTful API. To list records on an application level, you use the Content API (GET /contentapi/{level_alias}). To retrieve the nested details of a single record, you use the REST API (GET /api/core/content/{id}). Building an MCP server means you have to map LLM-friendly schemas to these split architectures and intelligently route the requests, otherwise Claude will constantly hallucinate endpoint structures.
Nested Field ID Payloads
When creating or updating an Archer record via the RESTful API, you cannot simply pass a flat JSON object like {"Title": "New Risk"}. Instead, Archer requires a FieldContents object keyed strictly by numeric Field IDs. A payload looks like {"FieldContents": {"123": {"Type": 1, "Value": "New Risk", "FieldId": 123}}}. To make matters worse, different data types require different inner structures. A Values List field needs ValuesListIds, a Cross-Reference needs ContentID, and a Date requires specific string formatting. An LLM cannot natively guess your custom instance's numeric IDs or specific type mappings without a heavily standardized schema layer.
Rate Limiting and Exhaustion
Archer instances are highly protective of their compute resources, particularly when running heavy OData queries or executing data feeds. When Archer's upstream API returns an HTTP 429 Too Many Requests, Truto passes that error directly to the caller. Truto does not retry, throttle, or absorb rate limit errors. Instead, it normalizes the upstream rate limit data into standard IETF headers (ratelimit-limit, ratelimit-remaining, ratelimit-reset). Your MCP client or agent framework is strictly responsible for implementing its own retry and exponential backoff logic.
Generating a Managed MCP Server for Archer
Instead of hardcoding tool definitions and managing session authentication lifecycles from scratch, you can use Truto to generate an MCP server dynamically. Truto translates Archer's actual API documentation into MCP-compliant JSON Schema representations, ensuring Claude always has the correct field types and parameter requirements.
You can create this server in two ways: through the Truto UI, or programmatically via the API.
Method 1: Via the Truto UI
If you are configuring this for internal team use, the easiest path is the dashboard.
- Log into your Truto account and navigate to the Integrated Accounts page.
- Select your connected Archer account.
- Click the MCP Servers tab.
- Click Create MCP Server.
- Select your desired configuration. You can filter tools by tags or specific HTTP methods (e.g., read-only).
- Click Generate and copy the resulting MCP Server URL (e.g.,
https://api.truto.one/mcp/abc123xyz).
Method 2: Via the Truto API
If you are provisioning MCP servers dynamically for your own customers, you can call Truto's management API. This validates that the connection is active, provisions a hashed token in a global KV store, and returns the secure URL.
curl -X POST https://api.truto.one/integrated-account/{integrated_account_id}/mcp \
-H "Authorization: Bearer YOUR_TRUTO_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"name": "Archer GRC Assistant",
"config": {
"methods": ["read", "write"],
"tags": ["grc", "security"]
}
}'This will return a JSON object containing the url. This URL contains a cryptographic token that routes requests directly to that specific Archer tenant. No additional OAuth logic is required on the client side.
Connecting the MCP Server to Claude
Once you have the Truto MCP URL, providing Claude with access to Archer takes less than a minute. You can do this via the Claude UI or via a configuration file for Claude Desktop.
Method 1: Via the Claude UI (Team/Enterprise)
If you are using Claude for Work (Team or Enterprise plans), you can add the server directly through the web interface.
- Go to Settings -> Integrations.
- Select Add MCP Server.
- Paste the Truto MCP URL and name the connection (e.g., "Archer Production").
- Click Add. Claude will instantly send an
initializerequest to discover the available Archer tools.
Method 2: Via the Claude Desktop Config File
For local development or individual Claude Desktop users, you connect the server via the claude_desktop_config.json file. Because Truto provides a standard Server-Sent Events (SSE) endpoint, you utilize the official MCP SSE client.
{
"mcpServers": {
"archer": {
"command": "npx",
"args": [
"-y",
"@modelcontextprotocol/server-sse",
"https://api.truto.one/mcp/YOUR_TRUTO_TOKEN"
]
}
}
}Restart Claude Desktop. You will see a plug icon indicating the tools have been successfully loaded.
Archer Hero Tools for Claude
Truto automatically derives dozens of tools from Archer's API definitions. Rather than dumping raw CRUD operations into the model's context window, these tools are highly opinionated and formatted to handle Archer's specific quirks.
Here are 6 high-leverage hero tools that transform Claude into a competent GRC administrator.
list_all_archer_records
Retrieves records from a specific application level using the Content API. Unlike the REST API, this endpoint allows for bulk retrieval (up to 1,000 records per page) and returns field values mapped to human-readable field aliases rather than numeric IDs. This is critical for LLMs, as they operate much better on semantic aliases than arbitrary numbers.
"Claude, use the list_all_archer_records tool to fetch all records from the 'Incidents_Risks' level and summarize the high-severity items from this week."
create_a_archer_record
Creates a new content record using the RESTful API. Truto's MCP schema specifically defines the FieldContents mapping requirements. It enforces the strict Type enumeration (e.g., 1 for Text, 3 for Date, 9 for Cross-Reference) so that Claude structures the nested payload exactly as Archer demands, dropping invalid inputs silently rather than crashing the request.
"Claude, create a new incident record in Archer. Set the title to 'Server Outage', the event date to today, and map it to cross-reference ID 4452."
list_all_archer_data_feed_recent_runs
Data feeds in Archer sync external data sources (like HR systems or vulnerability scanners) into the platform. This tool allows Claude to check the status of a specific feed's most recent run, differentiating between statuses like Running (1) and Pending (7).
"Claude, check the recent runs for the Vulnerability Scanner data feed and let me know if the latest execution failed or completed successfully."
create_a_archer_workflow_action
Advances a record through Archer's advanced workflow engine. Instead of a simple status update, Archer requires executing a transition. This tool explicitly requires the WorkflowNodeId and the specific CompletionCode (e.g., Approve, Reject, Route) retrieved from the list_all_archer_workflow_actions tool.
"Claude, review the current workflow actions available for content ID 8891, and execute the 'Approve' transition to move the policy to the next stage."
list_all_archer_security_events
Fetches the Security Events report for a specific day. This includes logins, password changes, and modifications to roles or groups. It supports heavy pagination (up to 50,000 events per page) and is essential for AI-driven security auditing and anomaly detection.
"Claude, pull the Archer security events for yesterday and flag any instances where a user's access role was changed to System Administrator."
create_a_archer_data_gateway_connection
Manages Data Gateway links to external systems. If Archer is relying on federated data without importing it into the local database, this tool configures the connection bridge, requiring a TargetId and a LocationAlias.
"Claude, add a new Data Gateway connection targeting the new regional database instance using the alias 'EU_West_Compliance_DB'."
To view the complete inventory of available Archer tools, schemas, and required parameters, visit the Archer integration page.
Workflows in Action
Providing an LLM with discrete tools is only the first step. The real value is realized when Claude chains these tools together to execute complex, multi-step GRC workflows autonomously.
Workflow 1: Auditing Security Events and Escalating Incidents
Security Operations Center (SOC) analysts spend hours correlating system access logs with recorded incidents. You can ask Claude to automate this audit trail analysis.
"Claude, audit yesterday's Archer security events for any unexpected role escalations. If you find any unauthorized changes to the 'System Administrator' group, check the recent runs of the HR Data Feed to see if the user was recently offboarded, and then create a high-severity incident record."
Execution flow:
- Claude calls
list_all_archer_security_eventspassing yesterday's date asEventsForDateand filters for role changes. - Upon finding an anomaly, Claude calls
list_all_archer_data_feed_recent_runsfor the HR feed to check the user's active status. - Claude uses
create_a_archer_recordto format the nestedFieldContentspayload, injecting the offending User ID and setting the severity field to 'High'. - Claude returns a summarized text report detailing the escalation and the ID of the newly created incident.
sequenceDiagram
participant User
participant Claude
participant MCP as Truto MCP
participant Archer as Archer API
User->>Claude: Audit events and escalate anomalies
Claude->>MCP: Call list_all_archer_security_events
MCP->>Archer: Query security logs
Archer-->>MCP: Return 50k log entries
MCP-->>Claude: Return logs
Claude->>MCP: Call create_a_archer_record
MCP->>Archer: POST /api/core/content with FieldContents
Archer-->>MCP: Return Content ID
MCP-->>Claude: Return success
Claude-->>User: Report createdWorkflow 2: Approving Compliance Exceptions via Advanced Workflow
Compliance teams often get bogged down manually moving records through stages once evidence is verified. Claude can be instructed to verify the state and transition the workflow.
"Claude, check the details of compliance exception record ID 10455. If the 'Risk Rating' is Medium or lower, advance the workflow action to 'Approved'."
Execution flow:
- Claude calls
get_single_archer_record_by_idpassing10455to retrieve the fullFieldContentspayload. - Claude parses the returned JSON, looking up the
Valueassociated with the 'Risk Rating' Field ID. - Confirming the risk is Medium, Claude calls
list_all_archer_workflow_actionsto get the currentWorkflowNodeIdand valid completion codes. - Claude calls
create_a_archer_workflow_actionpassing theContentId,WorkflowNodeId, and theCompletionCodefor approval. - Claude responds confirming the exception has been routed successfully.
Security and Access Control
Exposing a system of record like Archer to an autonomous AI agent introduces significant security considerations. Truto's managed MCP servers provide several layers of security to restrict what Claude can do:
- Method Filtering: You can restrict an MCP server to only allow
readoperations. If Claude attempts to write data, the server will reject it. You can also specify exact HTTP methods likegetorlist. - Tag Filtering: Limit the tool surface area by only exposing Archer resources tagged with specific labels (e.g.,
["security", "users"]). - Require API Token Auth: By setting
require_api_token_auth: trueduring server creation, the MCP server will demand a valid Truto API token in theAuthorizationheader of the client, ensuring the raw URL cannot be abused if leaked. - Time-to-Live (TTL): Set an
expires_atISO datetime. Once this threshold is reached, Truto automatically cleans up the server and destroys the cryptographic token in the edge KV store, revoking Claude's access immediately.
Strategic Wrap-Up
Connecting Claude to Archer's highly specific API architectures requires more than just formatting an OpenAPI spec. It requires a resilient middleware layer capable of translating semantic LLM requests into Archer's rigid, ID-driven payloads and advanced workflow states.
By leveraging Truto's dynamically generated MCP servers, you bypass the toil of managing custom API integration code, session lifecycles, and nested data parsing. Your engineering team can focus on designing high-leverage agentic workflows instead of fighting the Archer Content API.
FAQ
- Does Truto handle Archer rate limits automatically?
- No. When Archer returns an HTTP 429 rate limit error, Truto passes that error to the caller without retrying or applying backoff. It normalizes the upstream rate limit data into standard IETF headers, leaving the retry logic to the MCP client or agent framework.
- How do I restrict Claude to read-only access in Archer?
- When generating the MCP server via the Truto UI or API, set the method filter to `["read"]`. This ensures the MCP server only exposes GET and LIST operations, blocking Claude from creating, updating, or deleting Archer records.
- Can Claude trigger Archer data feeds?
- Yes. Using tools like `list_all_archer_data_feed_recent_runs` and `create_a_archer_data_feed_execution`, Claude can check the status of existing feeds and execute them directly via the MCP server.
- How does Claude handle Archer's complex Field ID mapping?
- Truto automatically translates Archer's API documentation into MCP-compliant JSON Schemas. When Claude uses a tool like `create_a_archer_record`, Truto's schema enforces the correct structure for the nested `FieldContents` payload, ensuring Claude provides the required Types, Values, and FieldIds.