MCP Server
Overview
Pensar provides a remote Model Context Protocol (MCP) server that lets AI tools interact directly with your Pensar workspace. Through MCP, you can list scans, launch pentests, review security issues, and retrieve auto-generated fixes using tools like Claude Code, Desktop, or Web, or even directly in Slack.
MCP is an open standard that allows AI assistants to connect to external tools and data sources. Pensar’s MCP server implements the Streamable HTTP transport so it works with any MCP-compatible client.
Prerequisites
Before connecting an MCP client to Pensar you need:
- A Pensar Console account at console.pensar.dev
- At least one workspace with a connected repository
- An MCP-compatible client (e.g., Claude Desktop, Claude Code, Cursor, Windsurf)
Server URL
The Pensar MCP server is available at:
Authentication
The Pensar MCP server uses OAuth 2.0 for authentication, following the MCP Authorization specification (2025-03-26). When you first connect, your MCP client will open a browser window where you sign in with your Pensar Console credentials. After signing in, the client receives a token and uses it automatically for all subsequent requests.
Most MCP clients handle the OAuth flow automatically. You do not need to create or manage API keys, just sign in.
Setup
Claude Code
Claude Desktop / Web
Cursor
Windsurf
- Open Claude Code by running
claudein your terminal - Add the Pensar MCP server:
- Run the
/mcpcommand, you’ll see Pensar with a needs authentication status - Select it and choose Authenticate to sign in with Pensar in your browser
Toolsets
A connection exposes every tool by default. To give an agent only some of them, name the toolsets you want, comma-separated, in a toolsets query parameter on the server URL:
If your client lets you set request headers, you can send the same list in an X-MCP-Toolsets header instead:
An unknown toolset name is rejected with a 400 error that lists the valid names.
Available Tools
The Pensar MCP server exposes the following tools to your AI coding agent:
Scans
list_scans
List all scans in your workspace.
Parameters: None
Returns: A list of scans with their IDs, labels, statuses, scan types, branches, and timestamps.
Example prompt: “Show me all my Pensar scans.”
get_scan
Get detailed information about a specific scan.
Parameters:
Returns: Detailed scan information including workspace name, status, error messages, issue count, and whether a report is ready.
Example prompt: “What’s the status of scan xyz-789?”
list_pentest_targets
List the targets of a pentest. Each target is an attack-surface endpoint the pentest exercised. Use a target ID to read its logs with list_target_logs or search_target_logs, even when the target produced no issue.
Parameters:
Returns: Each target’s ID, URL, application, status, attempt count, start and completion times, and any error. API keys require pentests:read.
dispatch_pentest
Queue a hosted pentest in the authenticated workspace. The scan runs asynchronously; an identical active launch may be reused.
Parameters:
Exact endpoint arguments:
Find discovered endpoint IDs with the list_endpoints or search_endpoints tools, pensar apps endpoints <appId> in the Apex CLI, or GET /apps/{appId}/endpoints in the Console API.
Exact lists use full depth and test the supplied endpoints without commit-based narrowing. Each selected application must be ready to test. Omit endpointIds to test eligible workspace applications at priority depth by default.
Invalid IDs or incompatible arguments return a tool error without expanding the requested scope.
Returns: The queued scan’s scanId, label, status, and confirmation message. Use get_scan with that ID to check completion. Failures are MCP tool errors (isError: true).
Example prompt: “Use dispatch_pentest with these two discovered endpoint IDs only: 7e4fe67a-61e5-4aa5-b93a-6871b470aaef and 59e9b6d5-d13b-4912-9e46-a1e801c1d053. Then check the returned scan ID until it completes.”
Issues
list_issues
List security issues with optional filtering.
Parameters:
Returns: A list of issues with their IDs, issueLabel references (VULN-000123), console urls, titles, severities, statuses, and file locations. Closed issues also carry closedAt, closedDisposition and duplicateOf, so a closed finding is not mistaken for a resolved one; all three are null while the issue is open.
Example prompt: “Show me all critical open issues.”
get_issue
Get detailed information about a specific security issue.
Parameters:
Returns: The issue object, including full CVSS, nested application, endpoint, and repository, and pullRequest (the tracked pull request URL, or null). API keys require issues:read.
Example prompt: “Give me the details on VULN-000123.”
get_issue_evidence
Read a finding’s PoC, execution output, evidence steps, attack path, and source snippet at its recorded commit.
Parameters: issueId (required), an issue UUID or label in your workspace.
Returns: The evidence object, including null reasons, redaction, and truncation flags. API keys require issues:read.
list_issue_retests
Read retest attempts for a finding, newest first.
Parameters: issueId (required), an issue UUID or label in your workspace.
Returns: The retest history. API keys require issues:read. This tool reads history; it does not launch a retest.
retest_issue
Queue a retest against a finding’s original target. Accepts an issueId containing an issue UUID or label such as VULN-000123 in your workspace.
After a target-scope failure, pass optional targetRetry: { domainId, targetUrl } to select a verified replacement for this run. The fields and validation are shared with Retest an issue in the Console API.
OAuth sessions and API keys with issues:write can launch retests. This is a write tool, so approve the launch in your client when prompted.
Returns: issueId, sessionId, status: "queued", and a message. Queued means the request was accepted, not that the finding is fixed. View the outcome in Console or read GET /issues/{issueId}/retests through the Console API.
Example prompt: “Retest VULN-000123 against its original target.”
update_issue
Update a security issue’s status, disposition, inventory endpoint association, or the pull request it tracks.
Parameters:
Pair closedDisposition with status: "closed" — without it the finding reads as Resolved whatever the reason says. Reopening clears it, and false-positive never carries one. other is not accepted, as it is being retired from the vocabulary.
A duplicate close must name its original in duplicateOf, another issue in the same workspace that is not itself closed as a duplicate, and no other disposition may carry one.
Endpoint updates validate workspace ownership, keep applicationId synchronized to the endpoint’s application, and reject duplicate (pentest, title, endpoint) associations.
Returns: The updated issue details, including close metadata, endpointId, applicationId, and pullRequest.
The change is also applied to the finding’s linked GitHub, Linear and Notion tickets.
Example prompt: “Close issue abc-123 as risk-accepted, we have a WAF rule in front of it.”
list_issue_pull_requests
List the pull requests linked to a security issue, with the status of each and how it was created.
Parameters:
Returns: Each pull request’s ID, URL, status, how it was created, and when. API keys require issues:read.
Fixes
list_fixes
List all available fixes for a security issue.
Parameters:
Returns: A list of fixes with their IDs and affected file paths.
Example prompt: “Are there any fixes for issue def-456?”
get_fix
Get detailed information about a specific fix, including the code diff.
Parameters:
Returns: The fix details including the file path, a unified diff of the changes, an explanation of what the fix does, and the related issue ID.
Example prompt: “Show me the diff for fix ghi-789.”
Agent Logs
list_agent_logs
List agent execution logs for a specific issue. Useful for understanding how the agent discovered and verified a vulnerability.
Parameters:
Returns: A list of agent log entries with timestamps, levels, roles, and content.
Example prompt: “Show me the agent logs for issue abc-123.”
search_agent_logs
Search agent execution logs by text pattern with configurable context lines.
Parameters:
Returns: Matching log entries with surrounding context.
Example prompt: “Search the agent logs for issue abc-123 for mentions of SQL injection.”
list_target_logs
List the agent execution logs from one pentest target, whether or not it produced an issue.
Parameters:
Returns: Log entries with their ID, timestamp, level, role, message, metadata, service, and subagent. API keys require pentests:read.
search_target_logs
Search the agent logs of one pentest target, with surrounding context entries.
Parameters:
Returns: Matching log entries with surrounding context. API keys require pentests:read.
Knowledge Base
Knowledge base entries are the context Pensar agents use during pentests; an entry’s priority decides when agents get it (see Knowledge Base). Entries steer what agents test and report, so the write tools need a user credential: an OAuth session can create, update, and archive entries, while a workspace API key can only read them. Creating, editing, or archiving an always entry also needs that user to be a workspace admin or owner. Every entry records the user who wrote it. Like dispatch_pentest, the write tools default to Needs approval.
list_knowledge
List knowledge base entries, newest first.
Parameters:
Returns: Entries with their ID, title, content, category, status, priority, source, tags, createdBy, and timestamps.
Example prompt: “What does Pensar know about our tech stack?”
get_knowledge
Get one knowledge base entry.
Parameters:
Returns: The full entry.
create_knowledge
Add a knowledge base entry, attributed to you. Requires a user credential.
Parameters:
Returns: The created entry.
Example prompt: “Add a technology note to Pensar that our API sits behind a WAF.”
update_knowledge
Edit a knowledge base entry. Omitted fields are unchanged; tags replaces the existing list. Requires a user credential.
Parameters:
Returns: The updated entry.
archive_knowledge
Archive a knowledge base entry so agents stop using it. The entry stays readable. Requires a user credential.
Parameters:
Returns: The archived entry.
Applications
list_applications
List the applications in your workspace, a page at a time.
Parameters:
Returns: A page of applications with their ID, name, type, framework, and domain, plus hasMore. When hasMore is true, pass the next offset to continue. API keys require inventory:read.
get_application
Get one application with its full detail.
Parameters:
Returns: The application, including its description and disallowed actions. API keys require inventory:read.
search_applications
Search applications by name and description.
Parameters:
Returns: A page of matching applications, plus hasMore. API keys require inventory:read.
create_application
Create an application.
Parameters:
Returns: The created application. API keys require inventory:write.
update_application
Update an application. Only the fields you pass change, and at least one is required.
Parameters:
Returns: The updated application. API keys require inventory:write.
delete_application
Permanently delete an application and all of its endpoints. This cannot be undone.
Parameters:
Returns: A confirmation with the deleted application’s ID. API keys require inventory:write.
Endpoints
list_endpoints
List the endpoints of one application, a page at a time.
Parameters:
Returns: A page of endpoints with their ID, path, transport, type, and whether authentication is required, plus hasMore. Objectives, business logic, and the threat model are left out; read them with get_endpoint or get_endpoints. API keys require inventory:read.
get_endpoint
Get one endpoint with its full detail.
Parameters:
Returns: The endpoint, including its description, objectives, source location, authentication details, business logic, and threat model. API keys require inventory:read.
get_endpoints
Get the full detail of up to 100 endpoints in one call.
Parameters:
Returns: The full detail of each endpoint, in the order requested. If any ID is not found in your workspace, the call fails with a 404 error and returns none of them. API keys require inventory:read.
search_endpoints
Search endpoints across your workspace, or within one application.
Parameters:
Returns: A page of matching endpoints, plus hasMore. Like list_endpoints, it leaves out objectives, business logic, and the threat model. API keys require inventory:read.
create_endpoint
Create a new endpoint under an application. If the application already has the same endpoint, the call returns a 409 error and changes nothing; it does not update the existing endpoint. Use update_endpoint to change it.
HTTP paths match after normalization, so case, a trailing slash, and parameter style such as :id or {id} do not matter. Other transports match on transport and exact path.
Parameters:
Returns: The created endpoint. API keys require inventory:write.
update_endpoint
Update an endpoint. Only the fields you pass change, and at least one is required.
Parameters:
A change that would give an application two endpoints with the same path and transport returns a 409 error.
Returns: The updated endpoint. API keys require inventory:write.
delete_endpoint
Permanently delete one endpoint. This cannot be undone.
Parameters:
Returns: A confirmation with the deleted endpoint’s ID. API keys require inventory:write.
Domains
list_domains
List the domains in your workspace.
Parameters: None
Returns: The domains with their ID, URL, and verification status. API keys require inventory:read.
create_domain
Add a domain to your workspace. Adding a domain does not verify it; see Verifying Domains.
Parameters:
Returns: The domain, and created, which is false when the domain was already in your workspace. API keys require inventory:add.
Console API
Every MCP tool except get_endpoints has a matching Console API endpoint for programmatic access via API keys. The Console API also exposes webhook endpoints, which are not available over MCP. See the Console API for authentication and the full endpoint reference.
Usage Examples
Once connected, you can interact with Pensar through natural language in your AI coding agent. Here are some common workflows:
Run a Security Scan
Ask your agent to launch a pentest and then check the results:
“Run a Pensar pentest and let me know when it finishes.”
The agent will use dispatch_pentest to start the scan, then get_scan to poll for completion. To select exact targets, provide discovered endpoint UUIDs and ask the agent to pass them as endpointIds; omitting that field uses the workspace launch behavior described above.
Review Security Issues
Ask your agent to summarize findings from a recent scan:
“Show me all high and critical issues from my latest scan.”
The agent will use list_scans to find the most recent scan, then list_issues with severity filters to retrieve the relevant findings.
Apply a Fix
Ask your agent to retrieve and apply a recommended fix:
“Get the fix for issue abc-123 and apply it to my codebase.”
The agent will use list_fixes to find available fixes, get_fix to retrieve the diff, and then apply the patch to your local files.
Supported Clients
The Pensar MCP server works with any client that supports the Streamable HTTP transport and OAuth-based authorization. Tested clients include Claude Code, Claude Desktop, Cursor, and Windsurf.
If your preferred client is not listed above, consult its documentation on adding a remote MCP server.
Next Steps
Incidents
Use the incidents toolset to investigate Datadog alerts. API keys need
incidents:read; OAuth connections use your workspace access.
list_incidentsreturns grouped alerts across all verdicts. Filter withverdict,status, ortriageStatus. PassnextCursorascursorfor the next page.list_incident_eventsreturns individual events. SetgroupOfto an incident id to inspect that alert group.get_incidenttakesincidentIdand returns the full raw payload, triage analysis, evidence, indicators, recommendations, and timestamps.list_incident_logstakesincidentIdand returns paginated execution parts for the current triage run. Restart pagination after re-triage or to refresh active tools.get_incident_log_attachmenttakesincidentIdandpartIdfrom a log entry withhasAttachment: true. Its download URL expires after 60 seconds and provides the full content stored outside the inline log.
These tools read data. Remediation text is a recommendation to review, not approval for an agent to execute it.