stepsecurity-mcp
OfficialThe StepSecurity MCP server exposes StepSecurity's platform APIs as LLM-callable tools for investigating supply-chain and CI/CD security issues in plain English.
Meta / Server Info
describe_capabilities– Self-describing summary of server capabilitiesping– Verify server reachabilityget_my_tenant– Confirm the configured StepSecurity tenantlist_tenant_github_orgs– List all GitHub orgs under your tenant
Threat Center
list_threat_incidents/get_threat_incident– List and detail supply-chain incidents, including compromised packages and C2 IOCs
Exposure & IOC Checks
Check if malicious npm/PyPI packages were used in CI or on developer machines
check_ioc_in_baseline– Search an org's Harden-Runner baseline for a domain/IP IOCfind_endpoint_calls_in_tenant– Tenant-wide search for workflows contacting a given endpointsearch_action_usage– Find which workflows use a specific GitHub Action (useful when one is reported compromised)
Detection Browsers (Tenant-Wide)
Browse anomalous network calls, blocked egress, HTTPS outbound calls, suspicious process events, imposter commit detections, secrets in build logs, and other detection types
Auditing & Inventory
list_recent_workflow_runs– Last 100 Harden-Runner-monitored runs for an org/repolist_github_api_calls_in_run– Audit every GitHub API call in a specific workflow run, grouped by job/stepfind_repos_using_endpoint– Find all repos in an org whose baseline contains a given endpoint
Suppression Rules (Read/Write)
List, get, preview, and analyze suppression rules
analyze_anomalous_calls_by_process– Group anomalous calls by process to identify suppression candidatesCreate, update, and delete suppression rules (write operations require user confirmation and an admin-scoped API key)
Allows checking if any workflow runs in the tenant contacted endpoints like bun.sh, useful for inventory and migration questions.
Enables analysis of anomalous network calls to identify Cloudflare WARP (cloudflared) traffic, aiding in suppression rule creation.
Provides tools to search for usage of specific GitHub Actions across repositories, audit GitHub API calls made by workflow runs, and list recent workflow runs.
Allows checking if a specific npm package is used across repositories and developer machines, useful for supply-chain incident response.
Enables analysis of anomalous network calls to identify OpenVPN traffic, aiding in suppression rule creation.
Allows checking if a specific PyPI package is used across repositories and developer machines, useful for supply-chain incident response.
Enables analysis of anomalous network calls to identify Tailscale VPN traffic, aiding in suppression rule creation.
Enables analysis of anomalous network calls to identify WireGuard traffic, aiding in suppression rule creation.
Enables analysis of anomalous network calls to identify ZeroTier VPN traffic, aiding in suppression rule creation.
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., "@stepsecurity-mcp Am I affected by the axios supply-chain incident?"
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.
stepsecurity-mcp
Model Context Protocol (MCP) server for the StepSecurity platform. Exposes a curated set of StepSecurity APIs as LLM-callable tools so you can investigate supply-chain and CI/CD security issues in plain English from Claude Desktop, Claude Code, Cursor, or any other MCP client.
Requirements
Node.js 22+
A StepSecurity API key
Your StepSecurity customer/tenant identifier (see Configure)
Related MCP server: armorcode-mcp
Install
No install needed for most users — wire the published package into your MCP client via npx (see Wire into an MCP client below). npx will fetch @stepsecurity/stepsecurity-mcp from npm on first run.
To hack on the server locally instead:
git clone git@github.com:step-security/stepsecurity-mcp.git
cd stepsecurity-mcp
npm install
npm run buildConfigure
Get an API key
Go to the StepSecurity admin console for your tenant and create a key:
https://app.stepsecurity.io/<tenant>/admin-console/integrations/stepsecurity-apiReplace <tenant> with your StepSecurity tenant name (the same value you'll put in STEP_SECURITY_CUSTOMER).
StepSecurity supports read-only tenant API keys — recommended for this MCP server since every tool shipped here is read-only (list / get / search). Use a read-only key unless you have a specific reason to grant write access.
Env vars
Variable | Required | Default | Purpose |
| yes | — | Your API key. Sent as |
| strongly recommended | — | Your tenant identifier. When set, |
| no |
|
|
Copy .env.example to .env for manual runs. For MCP clients, set the same values via the env block of the client config (next section).
Wire into an MCP client
Claude Code
Add to your project's .mcp.json (or user-level ~/.claude.json):
{
"mcpServers": {
"stepsecurity": {
"command": "npx",
"args": ["-y", "@stepsecurity/stepsecurity-mcp"],
"env": {
"STEP_SECURITY_API_KEY": "${STEP_SECURITY_API_KEY}",
"STEP_SECURITY_CUSTOMER": "your-tenant-name"
}
}
}
}If you're developing locally, swap command to node and args to ["/absolute/path/to/stepsecurity-mcp/dist/server.js"].
Restart, then run /mcp to confirm the server connected.
Claude Desktop
Edit ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) with the same shape, then restart Claude Desktop.
MCP Inspector (no-LLM debugging)
Fastest iteration loop — browse the tool list, fill args, see raw JSON-RPC:
STEP_SECURITY_API_KEY=... STEP_SECURITY_CUSTOMER=... npm run inspectOpens a web UI at http://localhost:6274.
Scenarios
The LLM chains multiple tool calls from one prompt — you don't need to know tool names. Ask any of the following in natural language:
1. Respond to a supply-chain incident
You heard about a malicious npm release or a compromised Action. Ask whether your tenant is affected. The LLM will: pull the incident from StepSecurity's threat center → read its "Am I Affected?" section → fan out to CI exposure, developer machines, and baseline IOC checks in parallel.
Try:
"Am I affected by the axios supply-chain incident?" (npm)
"Check exposure for the xinference PyPI compromise." (pypi)
"Check exposure for the trivy compromise." (compromised GitHub Action)
"There's a new malicious release mentioned on HN — anything in our threat center matching?"
Tools used behind the scenes: list_threat_incidents, get_threat_incident, and per-ecosystem exposure tools (check_npm_package_exposure + check_npm_package_on_dev_machines for npm, check_pypi_package_exposure + check_python_package_on_dev_machines for pypi), plus check_ioc_in_baseline / find_endpoint_calls_in_tenant for C2 indicators and search_action_usage for compromised Actions.
There's also a slash-prompt you can invoke explicitly in Claude Desktop: am-i-affected (args: org, incident) — expands into a guaranteed step-by-step investigation plan.
2. Browse recent detections in your tenant
Tenant-wide sweep of Harden-Runner detections across every org installed under your customer. Every detection returned includes a dashboard_url deep-link to the exact tab on the run page so you can click through to investigate.
Try:
"What are the recent detections in StepSecurity for my tenant?" — sweeps all six detection types in parallel.
"Show me any anomalous network calls."
"Were any secrets leaked in build logs?"
"Which domains got blocked this week?"
"Any imposter-commit detections I should look at?"
One tool per common detection type, all defaulting to tenant scope:
Tool | Detection type | Links to |
| New outbound calls off the baseline |
|
| Egress-policy blocks |
|
| HTTPS calls with method + path |
|
| Memory-read, reverse-shell, privileged container |
|
| Action pinned to a non-tag/non-branch SHA |
|
| Secret patterns in CI logs (masked) |
|
| Generic escape hatch for less-common types (source-code overwritten, harden-runner-config-changed, agent-tampered, etc.) | — |
3. Check a specific IOC
You have a domain or IP (e.g. from a threat feed) and want to know if any workflow contacted it.
Try:
"Was the C2 domain
sfrclak.comseen in our Harden-Runner baseline?" (single org)"Did any job across our tenant contact
registry.npmjs.org?" (tenant-wide)"Which workflow runs hit
142.11.206.73?"
Tools:
check_ioc_in_baseline(owner, indicator)— single org, server-side substring match. Each observation has adashboard_urllinking to the run's network-events tab.find_endpoint_calls_in_tenant(endpoint, customer?)— tenant-wide. Lists orgs then fans out per org with bounded concurrency. Returns a flat list of{org, repo, workflow, job, run_id, timestamp, dashboard_url}observations.
4. Audit usage of a specific GitHub Action
Triage when an Action is reported compromised.
Try:
"Which workflows use
aquasecurity/trivy-action?""Where is
tj-actions/changed-filesreferenced across our repos?"
Tool: search_action_usage (returns per-workflow pinning detail + upstream release info).
5. Audit GitHub API calls a workflow run made
You want to see every github.com / api.github.com call a specific run touched — catching writes to repos outside the org, surprising cross-org access, or API calls from third-party Actions.
Try:
"What GitHub API calls did the latest run of
trivy-scanin<org>/<repo>make?""Audit the API footprint of the most recent CI run in
<org>/<repo>.""For run 23336327796 of
<org>/<repo>, which GitHub API endpoints were called?"
Tools: list_recent_workflow_runs(owner, repo?) to find the run ID (when the user doesn't give one), then list_github_api_calls_in_run(owner, repo, runId) — single upstream call, grouped by job with step/tool attribution. The LLM will chain the two automatically if you phrase the question as "the latest run" or "yesterday's run".
6. Find every repo in an org contacting a given endpoint
You want to answer inventory / migration questions: "who still uses registry.npmjs.org?", "which repos contact bun.sh?".
Try:
"Which repos in
<org>contactbun.sh?""Find all repos still calling
registry.npmjs.org."
Tool: find_repos_using_endpoint(owner, endpoint, concurrency?) — fans out one baseline query per repo with bounded parallelism (default 10 in flight). Tens of seconds on large orgs. Returns only the matching repos, each with a baseline deep-link.
For a tenant-wide sweep across every org installed under your customer, first call list_tenant_github_orgs (takes customer optionally, falls back to STEP_SECURITY_CUSTOMER) to get the org list, then run find_repos_using_endpoint per org. The LLM will compose these when you ask something like "which repos across our tenant still use registry.npmjs.org?".
find_repos_using_endpoint(org-wide, repo-attributed) vscheck_ioc_in_baseline(single org, flat endpoint list): use the former when you want a per-repo inventory, the latter for quick IOC triage.
7. Triage detections and add suppression rules (write)
You want to analyze past detections — most often anomalous outbound network calls — decide which are false positives, and suppress them at the right scope. The LLM proposes rules, shows you the impact before anything writes, and only executes after you say yes.
Try:
"Analyze anomalous network calls across our tenant and suggest suppression rules for false positives."
"Suppress anomalous calls to
registry.npmjs.orgfrom thenpm-installjob in orgmy-org.""Show me all suppression rules configured for our tenant."
Tools (read):
analyze_anomalous_calls_by_process— tenant-wide: groups anomalies by calling process, flags VPN / mesh-networking daemons (tailscaled, twingate, zerotier-one, netbird, cloudflared, warp-svc, openvpn, wireguard) as strong candidates for a single process-scoped rule. One such rule matches both domain and direct-IP anomalies from that process (rule conditions withoutendpointorip_addressact as wildcards server-side). Other processes (dockerd, containerd, snapd, curl, kubelet, etc.) are returned too but not flagged — they deserve per-destination review.list_suppression_rules,get_suppression_rulepreview_suppression_rule(detectionId, conditions)— approximate client-side dry-run. Returns the detections that would be suppressed (count + sample with dashboard links).
Tools (write — require confirm: true and an admin-scoped API key):
create_suppression_rule— creates the rule. Server retroactively suppresses matching past detections synchronously; tool reports how many were moved.update_suppression_rule,delete_suppression_rule.
Never-suppress list:
gist.github.comandgist.githubusercontent.comare commonly used by attackers for payload delivery and exfiltration. The tools will refuse to create a rule that directly targets these hosts, andpreview_suppression_rulewill flag a warning if a broader rule would incidentally silence calls to them. Scope matters for auth: tenant-wide rules (conditions.owner = "*") require tenant admin. Org-level rules (owner = "<org>",repo = "*") require admin on that org. If you get403, the tool surfaces a clear message explaining to swap to an admin key or narrow the scope. Severity: onlyignore(pure suppression) is supported in the backend; the tool hardcodes it. Deletion doesn't un-suppress. Detections already moved to suppressed state stay that way after the rule is deleted.
8. Confirm your configured tenant
When in doubt about which tenant the tools are hitting.
Try:
"What's my tenant?"
"Which customer is this MCP server scoped to?"
Tool: get_my_tenant (reports the configured STEP_SECURITY_CUSTOMER and links to the tenant admin console).
Tool reference (all tools)
Grouped by area. Run describe_capabilities from the LLM for a self-describing summary.
Meta —
describe_capabilities,ping,get_my_tenantThreat center —
list_threat_incidents,get_threat_incidentExposure checks —
check_npm_package_exposure,check_npm_package_on_dev_machines,check_pypi_package_exposure,check_python_package_on_dev_machines,check_ioc_in_baseline,find_endpoint_calls_in_tenant,search_action_usageDetection browsers (tenant-wide) —
list_anomalous_network_calls,list_blocked_domain_calls,list_https_outbound_calls,list_suspicious_process_events,list_imposter_commit_detections,list_secrets_in_build_log,list_detectionsAuditing / inventory —
list_recent_workflow_runs,list_github_api_calls_in_run,find_repos_using_endpoint,list_tenant_github_orgsSuppression rules (write) —
list_suppression_rules,get_suppression_rule,preview_suppression_rule,analyze_anomalous_calls_by_process,create_suppression_rule,update_suppression_rule,delete_suppression_rule
Development
npm run dev # watch mode with tsx
npm run typecheck
npm run test
npm run buildProject layout
src/
server.ts MCP bootstrap (stdio transport) + server-level instructions
config.ts env loading (API key, customer, dashboard host)
logger.ts stderr-only JSON logger (stdout is reserved for JSON-RPC)
client.ts thin fetch wrapper with Bearer-auth injection
tools/
index.ts tool registry
describe.ts describe_capabilities
tenant.ts get_my_tenant
threat-center.ts incident + exposure-check tools
detections.ts per-type detection browsers (+ dashboard deep-links)
ping.ts
prompts/
index.ts am-i-affected slash-promptAvailable Tools
30 toolsanalyze_anomalous_calls_by_processA
Group tenant-wide anomalous network-call detections by the calling process. Goal: spot VPN / mesh-networking daemons (tailscaled, twingate, zerotier-one, netbird, cloudflared, warp-svc, openvpn, wireguard) that are legitimately fanning out to many peer IPs and coordination endpoints as normal operation. For those, a single process-scoped rule suppresses both domain AND direct-IP benign anomalies with one rule. Returns per-process: count, distinct endpoints, distinct direct IPs, sample detections (with dashboard links), and a suggested single suppression rule. When a VPN process appears (is_vpn_process_candidate=true), propose a process-wide rule (just {process: , owner: '*', ...}). Do NOT auto-propose process-wide rules for other processes (dockerd, containerd, snapd, curl, etc.) — those can make security-relevant calls and deserve per-destination review.
Inputs: customer, minCount.
check_ioc_in_baselineA
Search the Harden-Runner org baseline for a domain or IP indicator of compromise. Uses the server-side search query (case-insensitive substring) so only matching endpoints come over the wire. Returns which repos/workflows/runs contacted the endpoint; each observation has a dashboard_url — when presenting results you MUST include a clickable link per observation, not just the first one. For a tenant-wide search across every org under your customer, use find_endpoint_calls_in_tenant instead.
Inputs: owner (required), indicator (required).
check_npm_package_exposureA
Org-wide (CI-side) 'Am I Affected?' check for an npm package. Searches all monitored repos — workflow runs, PRs, and default branches — for uses of the package at the given versions. Pass an empty versions array to match any version. IMPORTANT: this only covers CI. For a complete 'am I affected' answer, also call check_npm_package_on_dev_machines in parallel to cover developer laptops.
Inputs: owner (required), packageName (required), versions, seenOnlyInPrs, +2 more.
check_npm_package_on_dev_machinesA
Developer-machine 'Am I Affected?' check for an npm package. Searches across all enrolled developer laptops (Dev Machine Guard) for installs of the package. Complements check_npm_package_exposure — CI and dev machines are INDEPENDENT exposure surfaces, so when investigating a malicious-package incident you MUST check both. Uses the StepSecurity customer/tenant identifier (optional — falls back to STEP_SECURITY_CUSTOMER env var). The server's version filter is not applied — pass versions to filter results client-side to specific compromised versions, otherwise all installs are returned.
Inputs: packageName (required), customer, versions.
check_pypi_package_exposureA
Org-wide (CI-side) 'Am I Affected?' check for a PyPI (Python) package. Searches all monitored repos — workflow runs, PRs, and default branches — for uses of the package at the given versions. Pass an empty versions array to match any version. IMPORTANT: this only covers CI. For a complete 'am I affected' answer, also call check_python_package_on_dev_machines in parallel to cover developer laptops. Use this (not check_npm_package_exposure) when the threat-center incident's ecosystem is 'pypi'.
Inputs: owner (required), packageName (required), versions, seenOnlyInPrs, +2 more.
check_python_package_on_dev_machinesA
Developer-machine 'Am I Affected?' check for a PyPI (Python) package. Searches across all enrolled developer laptops (Dev Machine Guard) for installs of the package. Complements check_pypi_package_exposure — CI and dev machines are INDEPENDENT exposure surfaces, so for a malicious-PyPI-package incident you MUST check both. The server's version filter is not applied — pass versions to filter results client-side to specific compromised versions, otherwise all installs are returned.
Inputs: packageName (required), customer, versions.
create_suppression_ruleA
Create a suppression rule. WRITE OPERATION — requires confirm: true and a read-only API key will 403. Before calling this, you MUST call preview_suppression_rule with the same conditions and show the user the expected impact. Creating a rule also retroactively suppresses matching past detections (synchronous server-side). After creation the tool verifies how many past detections were moved. Severity action is hardcoded to 'ignore' (only type the backend supports).
Inputs: detectionId (required), name (required), conditions (required), confirm (required), +2 more.
delete_suppression_ruleA
Delete a suppression rule. WRITE OPERATION — requires confirm: true. NOTE: deleting a rule does NOT un-suppress detections it previously matched; they remain in the suppressed state with the deleted rule_id attached.
Inputs: ruleId (required), confirm (required), customer.
describe_capabilitiesA
Describe what this MCP server can do and how to use it. Call this when the user asks 'what can you do?', 'what does this server support?', or is otherwise unsure how to start.
Inputs: no named parameters.
find_endpoint_calls_in_tenantA
Find every workflow-run observation of a given network endpoint across EVERY GitHub org installed under the tenant. Takes an endpoint substring (domain or IP), lists the tenant's orgs, and fans out a baseline search per org with bounded concurrency. Returns a flat list of observations: {org, repo, workflow, job, run_id, timestamp, dashboard_url}. Use this instead of check_ioc_in_baseline when the user asks 'did anyone in our tenant contact X?'. When presenting results you MUST include a clickable dashboard_url per observation.
Inputs: endpoint (required), customer, concurrency, observationsPerOrg.
find_repos_using_endpointA
Find every repo in an org whose Harden-Runner baseline contains a given network endpoint (domain or IP, substring match). Useful for questions like 'which repos contact bun.sh?', 'who still uses registry.npmjs.org?', or migration planning. Fans out one request per repo with bounded concurrency — expect 10–60 seconds on orgs with many repos. Returns only repos that matched — each one has a baseline_url which you MUST surface as a clickable link per repo (not just the first one). NOTE: this tool is single-org. For a tenant-wide sweep, first call list_tenant_github_orgs and then call this per org.
Inputs: owner (required), endpoint (required), concurrency.
get_my_tenantA
Return the StepSecurity customer/tenant identifier configured on this MCP server, along with a link to the tenant's admin console. Call this when the user asks 'what's my tenant?', 'which customer am I scoped to?', or wants to confirm the default before a detection sweep. Reads the STEP_SECURITY_CUSTOMER env var set in the MCP client config.
Inputs: no named parameters.
get_suppression_ruleB
Get one suppression rule by id. Read-only.
Inputs: ruleId (required), customer.
get_threat_incidentA
Get full details of one threat-center incident — including the 'Am I Affected?' markdown section which lists the EXACT compromised package names + versions and C2 IOCs (domains/IPs) to check. Call this after list_threat_incidents to extract the concrete things to search for. The incident's ecosystem field ('npm' or 'pypi') dictates which exposure tools to call: for npm → check_npm_package_exposure + check_npm_package_on_dev_machines. For pypi → check_pypi_package_exposure + check_python_package_on_dev_machines. Always also call check_ioc_in_baseline for each C2 domain/IP mentioned, regardless of ecosystem.
Inputs: owner (required), incidentId (required).
list_anomalous_network_callsA
List anomalous outbound network-call detections across the tenant (all orgs installed under the customer). 'Anomalous' = a destination endpoint was contacted that is NOT in the repo's Harden-Runner baseline of allowed endpoints — a common indicator of supply-chain exfiltration. Typically the most-used detection type during an investigation. Every result has a dashboard_url — when you present detections to the user you MUST include a clickable link per detection, not just the first one.
Inputs: customer, status, limit, orgScope.
list_blocked_domain_callsA
List detections where Harden-Runner actively BLOCKED an outbound network call (egress-policy enforcement). Different from anomalous calls: blocked = the call was prevented; anomalous = the call happened but wasn't in baseline. Every result has a dashboard_url — when you present detections to the user you MUST include a clickable link per detection, not just the first one.
Inputs: customer, status, limit, orgScope.
list_detectionsA
List Harden-Runner detections for an organization, filtered by detection type and status. Common detection IDs: 'Action-Uses-Imposter-Commit', 'Suspicious-Process-Events' (aggregates Runner-Worker-Memory-Read + Reverse-Shell + Privileged-Container), 'Anomalous-Outbound-Network-Call', 'Source-Code-Overwritten', 'Secret-In-Build-Log', 'Harden-Runner-Config-Changed', 'NPM-Package-Upgrade-To-Suspicious-Version', 'Agent-Tampered'.
Inputs: owner (required), detectionId (required), status, tenantWide.
list_github_api_calls_in_runA
List every HTTPS call to github.com or api.github.com made by jobs in a specific workflow run. Useful for auditing which GitHub API endpoints a workflow touches — detecting writes to unexpected repos, surprising org access, or API calls from third-party Actions. Fast: one upstream API call. Requires a run_id — if the user doesn't supply one, call list_recent_workflow_runs first to find it. Returns calls grouped by job with step/tool attribution. The response has a top-level dashboard_url for the run AND a per-job dashboard_url deep-linking to the network-events tab scoped to that job. When presenting results, include the per-job link next to each job header.
Inputs: owner (required), repo (required), runId (required).
list_https_outbound_callsA
List HTTPS outbound network-call detections (TLS-intercepted calls with method + path). Useful when you need to see WHAT an outbound call did — e.g. POSTs to a suspicious endpoint during a build. Every result has a dashboard_url — when you present detections to the user you MUST include a clickable link per detection, not just the first one.
Inputs: customer, status, limit, orgScope.
list_imposter_commit_detectionsA
List detections where a GitHub Action is pinned to a commit SHA that doesn't match any legitimate tag or branch head of that action's repo — a strong indicator of Action-tampering (e.g. a compromised tag pointing to malicious commit). Every result has a dashboard_url — when you present detections to the user you MUST include a clickable link per detection, not just the first one.
Inputs: customer, status, limit, orgScope.
list_recent_workflow_runsA
List the 100 most recent Harden-Runner-monitored workflow runs for a GitHub organization, optionally narrowed to one repository. Use this to discover run IDs when the user asks about a run without giving an explicit ID — e.g. 'the latest run of trivy-scan in poc-1'. Every result has a dashboard_url — when you present runs to the user you MUST include a clickable link per run, not just the first one.
Inputs: owner (required), repo, page.
list_secrets_in_build_logA
List detections where a secret (API key, private key, token, etc.) was detected in a CI build log. The API returns the secret already masked (e.g. '----****') — safe to display. Includes rule_id (which detector fired), line_number and step_number for navigation to the leak. Every result has a dashboard_url — when you present detections to the user you MUST include a clickable link per detection, not just the first one.
Inputs: customer, status, limit, orgScope.
list_suppression_rulesA
List all suppression (detection) rules configured for the tenant. Use this before creating a new rule to check for duplicates or near-overlaps. Read-only.
Inputs: customer.
list_suspicious_process_eventsA
List suspicious-process-event detections across the tenant. This is a virtual detection ID that aggregates three real types: Runner-Worker-Memory-Read (credential theft from runner memory), Reverse-Shell, and Privileged-Container. Use for runtime-evidence of compromise during an incident. Every result has a dashboard_url — when you present detections to the user you MUST include a clickable link per detection, not just the first one.
Inputs: customer, status, limit, orgScope.
list_tenant_github_orgsA
List every GitHub organization installed under a StepSecurity tenant. Call this first when a user asks for a tenant-wide view (e.g. 'find repos across my tenant using bun.sh') — then fan out find_repos_using_endpoint or similar per-org tools across the result. Each org has a dashboard_url pointing to its security summary — surface it as a clickable link per org. customer is optional; falls back to STEP_SECURITY_CUSTOMER.
Inputs: customer.
list_threat_incidentsA
List supply-chain threat-center incidents tracked by StepSecurity for a GitHub organization. ALWAYS call this first when the user mentions a known supply-chain incident or malicious package by name (e.g. 'axios compromise', 'trivy incident', 'Shai-Hulud') to find the matching incident and its ID before searching for exposure. Returns a compact summary (no markdown body) — then call get_threat_incident with the ID for full details.
Inputs: owner (required).
pingA
Returns 'pong' — use to verify the MCP server is reachable.
Inputs: message.
preview_suppression_ruleA
APPROXIMATE client-side preview of what creating a suppression rule would do. Fetches recent detections of the given detection_id and filters them against the proposed conditions, returning the count and up to 20 samples (with dashboard_url per sample). Use this BEFORE create_suppression_rule to show the user concrete impact. Preview uses exact + wildcard matching — the server's matcher is stricter (CIDR for ip_address, regex), so the final count after create may differ slightly.
Inputs: detectionId (required), conditions (required), customer.
search_action_usageA
Find which workflows across the organization use a given GitHub Action. Useful for responding to a compromised Action (e.g. 'which repos use aquasecurity/setup-trivy?'). Returns one entry per workflow that references the Action.
Inputs: owner (required), action (required).
update_suppression_ruleA
Update an existing suppression rule (name, description, or conditions). WRITE OPERATION — requires confirm: true.
Inputs: ruleId (required), confirm (required), customer, name, +2 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 30 tools
Tools are generally distinct but the many detection listing tools (e.g., list_anomalous_network_calls, list_blocked_domain_calls, list_https_outbound_calls, list_suspicious_process_events) could cause confusion despite descriptive names. Overlap is minimal due to specific filters.
All tool names follow a consistent verb_noun snake_case pattern (e.g., list_anomalous_network_calls, create_suppression_rule). No mixing of conventions or abbreviations, making the pattern predictable.
30 tools is high but justified for a security monitoring server covering incident response, suppression rules, detection listing, and package exposure. The scope is broad, but each tool has a clear purpose and the count is not excessive given the domain.
The tool surface covers detection management, threat incident analysis, package exposure (CI and dev machines), and suppression rules. Minor gaps exist, such as a generic 'list all detections' tool or user management, but core workflows are well-covered.
Maintenance
Related MCP Connectors
Security scanner for MCP servers. Detect vulnerabilities, prompt injection, and tool poisoning.
An MCP server for Arcjet - the runtime security platform that ships with your AI code.
MCP server for Appcircle mobile CI/CD platform.
MCP server for secureFlows: token-free URL builders and integration-linting tools for AI agents.
Related MCP Servers
- AlicenseAqualityCmaintenanceMCP server for Cursor that scans codebases for security issues including hardcoded secrets, SAST, vulnerable dependencies, and IaC misconfigurations.7MIT
- AlicenseCqualityDmaintenanceMCP server for ArmorCode security platform enabling finding triage, scans, exceptions, risk scores, and release gate checks via natural language.3731 npmMIT
- AlicenseNot gradedqualityBmaintenanceStandalone MCP server that provides security scanning, project mapping, and vulnerability fix generation to AI coding assistants.1,118 npm2MIT
- FlicenseNot gradedqualityDmaintenanceMCP server for Legit Security that answers questions about security issues, Actions, inventory, posture, and platform usage across one or multiple tenants.-