n0s1-mcp
This server exposes secret-scanning capabilities as MCP tools, allowing AI assistants to scan various platforms and filesystems for leaked credentials and secrets.
Scanning
Jira — Scan tickets for leaked secrets, with JQL query scoping and optional auto-posting of warning comments
Confluence — Scan pages using CQL query scoping
Slack — Scan channels using a bot token
GitHub — Scan repositories with optional branch/repo filtering
GitLab — Scan projects, including self-hosted instances
Zendesk — Scan tickets for leaked secrets
Linear — Scan issues for leaked secrets
Asana — Scan tasks with optional workspace/project scoping
Wrike — Scan tasks with optional folder/space scoping
Local filesystem — Scan a local path, with support for custom regex rules
Monitoring & Analysis
Check scan status — Monitor progress of a running scan via its UUID
Get scan findings — Retrieve paginated, severity-filterable findings from completed scans
Analyze reports — Queue async AI-powered credential validation on scan results (requires n0s1 Professional)
Other features: Multiple report formats (n0s1, sarif, gitlab), credentials passable as tool arguments or environment variables, and custom UUID assignment for scan reports.
Scan Asana tasks for leaked secrets, including task details and attachments.
Scan Confluence pages for leaked secrets, including page content and attachments.
Scan GitHub repositories for leaked secrets, including code, issues, and pull requests.
Scan GitLab projects for leaked secrets, including code, issues, and merge requests.
Scan Jira tickets for leaked secrets using the n0s1 engine, including ticket contents and metadata.
Scan Linear issues for leaked secrets, including issue descriptions and comments.
Scan Slack channels for leaked secrets, including messages and files.
Scan Zendesk tickets for leaked secrets, including ticket comments and attachments.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@n0s1-mcp Scan the GitHub org mycompany for exposed API keys"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
n0s1-mcp
An MCP server that exposes n0s1 secret-scanning capabilities as tools for AI assistants (Claude, Cursor, etc.).
Scan Jira, Confluence, Slack, GitHub, GitLab, Zendesk, Linear, Asana, Wrike, and local filesystems for leaked secrets — directly from your AI workflow.
Quickstart
No install required. Add this to your MCP client config and run via uvx:
{
"mcpServers": {
"n0s1": {
"command": "uvx",
"args": ["n0s1-mcp"]
}
}
}For Claude Desktop: ~/Library/Application Support/Claude/claude_desktop_config.json
For Claude Code: .claude/mcp.json in your project, or ~/.claude/mcp.json globally.
Related MCP server: Offensive360 MCP Server
Available Tools
Tool | Description | Required params |
| Scan Jira tickets |
|
| Scan Confluence pages |
|
| Scan Slack channels |
|
| Scan GitHub repositories |
|
| Scan GitLab projects |
|
| Scan Zendesk tickets |
|
| Scan Linear issues |
|
| Scan Asana tasks |
|
| Scan Wrike tasks |
|
| Scan local filesystem |
|
| Get status of a running/completed scan |
|
| Get paginated findings for a completed scan |
|
| Submit or advance async AI credential validation |
|
All scan_* tools accept these optional parameters:
Parameter | Description |
| UUID to assign to the scan report. When set, overrides the auto-generated UUID written to the report JSON. |
| Queue async AI credential validation after the scan (requires n0s1 Professional) |
| n0s1 API key; overrides |
| Allow encrypted secrets to be uploaded to the n0s1 backend (default: |
Pass wait_minutes to analyze_report (or directly on a scan tool alongside ai_analysis) to block until analysis completes.
Environment Variables
Credentials can be passed as tool arguments or pre-set as environment variables:
Variable | Used by |
| All scan tools with |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Example with env vars pre-configured:
{
"mcpServers": {
"n0s1": {
"command": "uvx",
"args": ["n0s1-mcp"],
"env": {
"GITHUB_TOKEN": "ghp_...",
"JIRA_TOKEN": "ATATT..."
}
}
}
}Usage Examples
Once connected, ask your AI assistant:
"Scan my Jira project SEC for leaked secrets"
"Check the GitHub org mycompany for exposed API keys"
"Scan the /home/user/project directory for secrets"
"Run an AI analysis on the scan report abc123"
For full parameter reference and AI analysis workflow details, see docs/ai.md.
Publishing to PyPI
pip install hatch
hatch build
hatch publishLicense
GNU General Public License v3 — same as n0s1.
Available Tools
13 toolsanalyze_reportA
Submit or advance async AI credential validation for a previously uploaded scan report. Side effects: sends live HTTP validation requests to check whether discovered credentials are still active — this contacts the services where the secrets were found. Auth: requires n0s1_api_key or N0S1_TOKEN env var (n0s1 Professional account). Call once to queue, then poll until ai_analysis_status is 'complete' or 'failed'. Pass report_file when status is 'waiting_client' to inject credentials into validators. Pass wait_minutes to block until a terminal state or timeout; returns ai_analysis_status='timeout' if the deadline is reached without completion.
Inputs: report_uuid (required), n0s1_api_key, report_file, wait_minutes. Returns: Result returned by analyze_report.
get_scan_findingsARead-only
Return a paginated list of findings for a completed scan. Read-only with no side effects. All secret values are redacted — raw secrets are never returned. Pass next_cursor from a previous response to retrieve subsequent pages.
Inputs: report_uuid (required), page, severity. Returns: Paginated findings list for an existing scan.
get_scan_statusARead-only
Return the current status of a previously started scan. Read-only with no side effects — queries in-process scan state only. Returns 'pending' if the report_uuid is not yet known.
Inputs: report_uuid (required). Returns: Lightweight status check for an existing scan.
scan_asanaA
Read Asana tasks and comments to detect leaked secrets (API keys, tokens, passwords). Never modifies Asana — no tasks or comments are written. Auth: requires an Asana personal access token; set ASANA_TOKEN env var or pass api_key directly. Side effects: a redacted scan report is uploaded to the n0s1 backend; set allow_secret_upload=True to also upload AES-encrypted secret values for AI validation. Returns redacted findings — raw secret values are never included in the output. Subject to Asana API rate limits.
Inputs: api_key (required), scope, report_format, show_matched_secret_on_logs, +4 more. Returns: Full result returned by every scan_* tool.
scan_confluenceA
Read Confluence pages and comments to detect leaked secrets (API keys, tokens, passwords). Never modifies Confluence — no pages or comments are written. Auth: requires CONFLUENCE_TOKEN (falls back to JIRA_TOKEN) and CONFLUENCE_EMAIL (falls back to JIRA_EMAIL) env vars, or pass api_key/email directly. Side effects: a redacted scan report is uploaded to the n0s1 backend; set allow_secret_upload=True to also upload AES-encrypted secret values for AI validation. Returns redacted findings — raw secret values are never included in the output. Subject to Confluence API rate limits.
Inputs: server (required), email (required), api_key (required), scope, +6 more. Returns: Full result returned by every scan_* tool.
scan_githubA
Read GitHub repository code, issues, and pull requests to detect leaked secrets. Never modifies GitHub — no commits, comments, or PRs are created. Auth: requires a personal access token with repo (or public_repo) scope; set GITHUB_TOKEN env var or pass api_key directly. Side effects: a redacted scan report is uploaded to the n0s1 backend; set allow_secret_upload=True to also upload AES-encrypted secret values for AI validation. Returns redacted findings — raw secret values are never included in the output. Subject to GitHub API rate limits (5,000 req/hr authenticated).
Inputs: api_key (required), owner (required), repo, branch, +7 more. Returns: Full result returned by every scan_* tool.
scan_gitlabA
Read GitLab project code, issues, and merge requests to detect leaked secrets. Never modifies GitLab — no commits, comments, or MRs are created. Auth: requires a personal access token with read_api scope; set GITLAB_TOKEN env var or pass api_key directly. Side effects: a redacted scan report is uploaded to the n0s1 backend; set allow_secret_upload=True to also upload AES-encrypted secret values for AI validation. Returns redacted findings — raw secret values are never included in the output. Subject to GitLab API rate limits.
Inputs: api_key (required), owner (required), server, repo, +7 more. Returns: Full result returned by every scan_* tool.
scan_jiraA
Read Jira tickets and comments to detect leaked secrets (API keys, tokens, passwords). Never modifies Jira — no comments are posted, no tickets are changed. Auth: requires JIRA_TOKEN and JIRA_EMAIL env vars, or pass api_key/email directly. Side effects: a redacted scan report is uploaded to the n0s1 backend; set allow_secret_upload=True to also upload AES-encrypted secret values for AI validation. Returns redacted findings — raw secret values are never included in the output. Subject to Jira API rate limits.
Inputs: server (required), email (required), api_key (required), scope, +6 more. Returns: Full result returned by every scan_* tool.
scan_linearA
Read Linear issues and comments to detect leaked secrets (API keys, tokens, passwords). Never modifies Linear — no issues or comments are written. Auth: requires a Linear personal API key (lin_api_...); set LINEAR_TOKEN env var or pass api_key directly. Side effects: a redacted scan report is uploaded to the n0s1 backend; set allow_secret_upload=True to also upload AES-encrypted secret values for AI validation. Returns redacted findings — raw secret values are never included in the output. Subject to Linear API rate limits.
Inputs: api_key (required), report_format, show_matched_secret_on_logs, ai_analysis, +3 more. Returns: Full result returned by every scan_* tool.
scan_localARead-only
Scan a local filesystem path for leaked secrets (API keys, tokens, passwords). Fully local — no network calls, no data sent to any external service. Never modifies scanned files. No authentication required. Returns redacted findings — raw secret values are never included in the output.
Inputs: scan_path (required), regex_file, report_format, show_matched_secret_on_logs, +1 more. Returns: Full result returned by every scan_* tool.
scan_slackA
Read Slack messages to detect leaked secrets (API keys, tokens, passwords). Never modifies Slack — no messages are posted or edited. Auth: requires a bot token with channels:history and channels:read scopes; set SLACK_TOKEN env var or pass api_key directly. Side effects: a redacted scan report is uploaded to the n0s1 backend; set allow_secret_upload=True to also upload AES-encrypted secret values for AI validation. Returns redacted findings — raw secret values are never included in the output. Subject to Slack API rate limits.
Inputs: api_key (required), report_format, show_matched_secret_on_logs, ai_analysis, +3 more. Returns: Full result returned by every scan_* tool.
scan_wrikeA
Read Wrike tasks and comments to detect leaked secrets (API keys, tokens, passwords). Never modifies Wrike — no tasks or comments are written. Auth: requires a Wrike permanent access token; set WRIKE_TOKEN env var or pass api_key directly. Side effects: a redacted scan report is uploaded to the n0s1 backend; set allow_secret_upload=True to also upload AES-encrypted secret values for AI validation. Returns redacted findings — raw secret values are never included in the output. Subject to Wrike API rate limits.
Inputs: api_key (required), scope, report_format, show_matched_secret_on_logs, +4 more. Returns: Full result returned by every scan_* tool.
scan_zendeskA
Read Zendesk tickets and comments to detect leaked secrets (API keys, tokens, passwords). Never modifies Zendesk — no tickets or comments are written. Auth: requires ZENDESK_TOKEN, ZENDESK_EMAIL, and ZENDESK_SERVER env vars, or pass server/email/api_key directly. Side effects: a redacted scan report is uploaded to the n0s1 backend; set allow_secret_upload=True to also upload AES-encrypted secret values for AI validation. Returns redacted findings — raw secret values are never included in the output. Subject to Zendesk API rate limits.
Inputs: server (required), email (required), api_key (required), report_format, +5 more. Returns: Full result returned by every scan_* tool.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 13 tools
Each tool targets a distinct platform or function (scan, status, findings, analysis), with no overlap. The scan_* tools are clearly separated by platform name.
All tools follow a consistent verb_noun pattern in snake_case, e.g., scan_github, get_scan_status. No mixed conventions.
13 tools is well-scoped for a secret scanning server covering multiple integrations plus post-scan operations. Each tool has a clear purpose.
Covers scanning, status, findings, and analysis for many platforms. Missing tools for listing/deleting past scans or managing reports, but core workflow is complete.
Maintenance
Related MCP Connectors
Scan configs, files, or text for leaked secrets and obvious misconfigurations. Nothing stored.
Security scanner for AI agents: prompt injection, leaked secrets, PII, and SSRF-risk URL detection.
Scan websites and code for security, performance and SEO issues from your AI assistant.
Security scanner for AI-written code: SAST, secrets, dependencies, IaC, containers and DAST.
Related MCP Servers
- AlicenseNot gradedqualityCmaintenanceEnables AI assistants to scan source code for secrets like API keys and passwords locally without network requests.2Apache 2.0
- AlicenseAqualityDmaintenanceEnables AI assistants to run Offensive360 SAST scans on local codebases, returning security findings with file/line, severity, and fixes, plus scan status tracking.243 npmMIT
- AlicenseNot gradedqualityBmaintenanceEnables AI agents to scan and audit environment files for leaked secrets, sanitize them, encrypt/decrypt vaults, and run processes with zero-disk secret injection.MIT
- AlicenseNot gradedqualityBmaintenanceEnables AI agents to scan code or text content for hardcoded API keys, passwords, tokens, and private keys, reporting each finding by type, confidence, key name, and position. Secret values are never returned, so credentials can be located and addressed without leaving the machine.1MIT