MCP Airlock
MCP Airlock is a zero-trust security gateway that enforces capability-based access control for MCP tool calls, with tamper-evident audit logging.
Issue Capability Leases (
airlock_issue_capability): Request short-lived, HMAC-signed leases bound to a specific session, subject (agent), intent, and set of tools — with optional constraints (e.g., allowed domains, max risk) and a custom TTL (default 900s). These leases must be presented with subsequent tool calls to authorize execution.Audit Trail Inspection (
airlock_audit_tail): Read recent signed provenance events from a tamper-evident, hash-chained ledger to review allow/deny decisions (configurable limit, default 20 events).Fetch External JSON (
http_get_json): Make SSRF-resistant HTTP GET requests to public HTTPS endpoints with optional query parameters, subject to domain allowlisting and capability/policy checks.Hourly Weather Lookup (
weather_hourly): Resolve a US city/state to coordinates and fetch hourly forecasts from Open-Meteo (1–168 hours ahead, default 24), gated behind capability and policy enforcement.Dynamic Policy Enforcement: Every tool call is validated against risk thresholds, session/intent continuity, and tool-specific constraints — preventing prompt injection and privilege escalation.
Tamper-Evident Provenance: All authorization decisions are recorded in an append-only hash chain, ensuring auditability and integrity of the security log.
Supports OpenTelemetry traces and SIEM sinks for monitoring and observability of security decisions and tool calls.
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., "@MCP Airlock issue a short-lived capability for weather lookup"
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.
MCP Airlock
Zero-trust security gateway for MCP tools.
MCP Airlock turns every tool call into a short-lived, context-bound capability decision with tamper-evident provenance.
Why this exists
Agent tool ecosystems are failing at one painful boundary: the jump from untrusted prompt text to privileged tool execution.
Current patterns are usually one of:
static allowlists (
agent can call tool X)weak regex filtering
post-hoc logs with no integrity guarantees
They fail when prompt injection mutates intent mid-session, causing silent privilege escalation or data exfiltration.
MCP Airlock solves this with a missing primitive for MCP:
Capability Leases: short-lived, signed, context-bound rights (
session + intent + tool scope + constraints)Context-Aware Policy: dynamic authorization on every call (risk score + tool constraints + lease checks)
Tamper-Evident Provenance: append-only hash chain across all allow/deny decisions
Related MCP server: mcp-permission-server
Core innovation
Context-Bound Capability Leases (CBCL)
Each tool call is authorized against a signed lease:
Bound to
session_idBound to
intent_hashScoped to specific tools
Time-limited
Optional constraints (e.g. allowed domains, max risk)
If a prompt injection tries to change intent or jump tool scope, execution is denied.
Architecture
flowchart LR
A[Agent / MCP Client] -->|tools/call| B[MCP Airlock Server]
B --> C[Risk Engine]
B --> D[Capability Verifier]
B --> E[Policy Engine]
E -->|allow| F[Tool Adapter Layer]
E -->|deny| G[Policy Deny Response]
F --> H[External APIs / Internal Services]
B --> I[Provenance Ledger Hash Chain]Trust boundaries
flowchart TB
subgraph Untrusted
U1[Prompt Content]
U2[Agent Reasoning Trace]
end
subgraph Trusted Control Plane
T1[MCP Airlock]
T2[Policy + Lease Validation]
T3[Signed Provenance Ledger]
end
subgraph External Targets
X1[Public APIs]
X2[Internal APIs]
end
U1 --> T1
U2 --> T1
T1 --> T2
T2 --> X1
T2 --> X2
T1 --> T3Key features
MCP stdio server compatible with
initialize,tools/list,tools/callCapability issuance tool:
airlock_issue_capabilityAgent/API usage analytics tool:
airlock_usage_statsAPI exposure measurement tool:
airlock_exposure_reportPolicy enforcement middleware with per-tool risk thresholds
Prompt-injection signature scoring
SSRF-resistant HTTP tool adapter (
http_get_json)Real API integration example (
weather_hourly)Tamper-evident provenance log + verification command
Sandbox hardening guide for agentic API security
CLI for serve/demo/issue/verify/stats/exposure
2-minute quickstart
git clone https://github.com/lara-muhanna/mcp-airlock
cd mcp-airlock
python -m pip install -e .
python -m mcp_airlock --config examples/airlock.config.json demoWhat you will see:
a human-friendly summary (handshake, capability, allow/deny, audit integrity)
malicious call denied with plain-English reasons
signed provenance evidence
One-command local demo (no install)
python -m mcp_airlock --config examples/airlock.config.json demo --city Austin --state TexasFor full JSON payloads during demo:
python -m mcp_airlock --config examples/airlock.config.json demo --rawRun as MCP server
python -m mcp_airlock --config examples/airlock.config.json serveMCP client setup examples:
CLI
# Issue a capability directly
python -m mcp_airlock --config examples/airlock.config.json issue \
--session-id sess-123 \
--subject agent:planner \
--tools weather_hourly,http_get_json \
--intent "Plan safe outdoor activities" \
--ttl-seconds 900 \
--constraints '{"allowed_domains":["api.open-meteo.com"],"max_risk":0.6}'
# Verify audit integrity
python -m mcp_airlock --config examples/airlock.config.json verify-log
# API usage stats by agent
python -m mcp_airlock --config examples/airlock.config.json stats --lookback-hours 24
# API exposure measurement
python -m mcp_airlock --config examples/airlock.config.json exposure --lookback-hours 24Example agent integration
Run:
python examples/agent_integration.pyThis script:
starts Airlock over stdio
negotiates MCP initialize/list
issues a lease
runs a normal tool call
runs an injected call that gets blocked
Config template
examples/airlock.config.json
{
"secret_key": "dev-secret-change-this-before-production",
"provenance_log": "./airlock-provenance.log",
"max_ttl_seconds": 1800,
"default_risk_threshold": 0.55,
"tools": {
"weather_hourly": {
"require_capability": true,
"risk_threshold": 0.7
},
"http_get_json": {
"require_capability": true,
"risk_threshold": 0.45,
"allowed_domains": ["api.open-meteo.com", "geocoding-api.open-meteo.com"]
}
}
}Security model summary
Agent requests lease via
airlock_issue_capability.Lease is HMAC-signed and includes
session,intent_hash,tool_scope,expiry.Every
tools/callrequest includes_capabilityand_context.Airlock enforces:
lease validity + signature
session and intent continuity
risk threshold
tool-specific constraints (e.g., domain allowlist)
Decision + evidence is hash-chained to provenance log.
Project structure
mcp-airlock/
mcp_airlock/
cli.py
server.py
policy.py
capability.py
risk.py
provenance.py
config.py
tool_ids.py
tools/
http_json.py
weather.py
examples/
airlock.config.json
agent_integration.py
docs/
CLIENT_SETUP.md
SANDBOXING_AGENTIC_APIS.mdRoadmap
Upstream MCP proxy mode (wrap existing MCP servers transparently)
OPA/Rego policy backend
OpenTelemetry traces + SIEM sinks
Managed capability broker + key rotation
Signed replay package for incident response
Community
Contribution guide:
CONTRIBUTING.mdSecurity policy:
SECURITY.mdCode of conduct:
CODE_OF_CONDUCT.mdRelease checklist:
docs/RELEASE_CHECKLIST.md
License
MIT
Available Tools
4 toolsairlock_audit_tailC
Read recent signed provenance events.
Inputs: limit.
airlock_issue_capabilityC
Issue a short-lived capability lease bound to session+intent.
Inputs: session_id (required), subject (required), tools (required), intent (required), +2 more.
http_get_jsonA
Fetch JSON from a public HTTPS API endpoint.
Inputs: url (required), query.
weather_hourlyA
Resolve US city/state and fetch hourly weather from Open-Meteo.
Inputs: city (required), state (required), hours.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 4 tools
Each tool targets a distinct function (provenance auditing, capability issuance, generic HTTP fetching, and weather retrieval) with minimal overlap. An agent can easily distinguish when to use each based on the task requirements.
Mixed naming patterns: two tools use 'airlock_' prefix with verbs (issue, audit_tail), while others use 'http_' prefix or no prefix (weather_hourly). The weather tool breaks the verb-first convention used by others, using a noun-adjective pattern instead.
Four tools is acceptably compact but the set suffers from scope confusion, mixing core Airlock security functions with unrelated utility tools (weather, HTTP). This feels like two different servers merged together rather than a cohesive toolset.
The Airlock domain (capability management) is severely underdeveloped, offering only issuance and audit tail reading without revocation, validation, or capability listing. The utility tools (weather, HTTP) are also minimal, offering only hourly forecasts and GET requests without parameter customization.
Maintenance
Related MCP Connectors
MCP gateway with runtime security policy, tool-call-level control, and audit of agent actions.
Find, vet, and run MCP tools through a secure audited gateway with prompt-injection risk scoring
Zero-secret MCP gateway for AI agents: risk-scored, audited calls with human-in-the-loop approval.
Identity, authorization, audit trails, and revocable permissions for AI agents accessing MCP tools.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceEnables secure interaction between LLMs and MCP tools by applying zero-trust security controls, including sensitive data masking, file system protection, and policy enforcement.MIT
- AlicenseNot gradedqualityCmaintenanceEnforces fine-grained, context-aware access control on MCP tool calls, with a tamper-evident, replayable audit log that records denials and verifies every decision.MIT
- FlicenseNot gradedqualityCmaintenanceEnables LLM agents to enforce security policies by authorizing tool calls, scanning for prompt injection, checking memory writes, redacting PII, and accessing audit trails through MCP.-
- AlicenseNot gradedqualityCmaintenanceEnables production-hardened MCP tool serving with OAuth 2.1 audience-restricted tokens, purpose-bound tools, human-in-the-loop approvals, containment, and audit trails for AI agents.Apache 2.0