feat: Add rate limiting and quota management for agent operations

Implements per-agent and per-user rate limiting with configurable quotas
for API calls, VM spawning, and approval requests.

Features:
- Token bucket algorithm for smooth rate limiting
- Per-agent and per-user quota tracking
- Configurable limits for different operation types
- Admin dashboard support for monitoring and adjusting quotas
- Admin user exemptions from rate limits

Components:
- RateLimitConfig: Configuration with environment variable support
- Quota: Token bucket implementation with refill logic
- QuotaStore: In-memory storage for quota data
- RateLimitManager: Central manager for rate limit checks
- DashboardData: Admin dashboard support

Tests: 43 unit tests covering all components

Closes: luminaguard-548

Author: Auto Merge Bot

.beads/issues.jsonl

@@ -27,7 +27,7 @@
 {"id":"luminaguard-4aj","title":"Implement Hackable Installation Mode","description":"Part of #440 Daemon Mode epic. Implement source code installation option with git clone and build, developer-friendly debugging, hot-reload, and extensible architecture for custom modifications","status":"closed","priority":1,"issue_type":"task","created_at":"2026-02-17T14:51:32.693165703-05:00","created_by":"alexc","updated_at":"2026-02-17T14:54:57.846922126-05:00","closed_at":"2026-02-17T14:54:57.846922126-05:00","close_reason":"Implemented hackable installation mode with hot-reload, extensible architecture, and comprehensive developer documentation","external_ref":"gh-454"}
 {"id":"luminaguard-4k9","title":"Requirements","description":"- Async server using asyncio to run all connectors","status":"closed","priority":2,"issue_type":"task","created_at":"2026-02-16T20:01:45.14238863-05:00","updated_at":"2026-02-16T20:02:19.96568719-05:00","closed_at":"2026-02-16T20:02:19.96568719-05:00","close_reason":"Closed"}
 {"id":"luminaguard-4sm","title":"Dependencies","description":"- discord.py\u003e=2.0","status":"closed","priority":2,"issue_type":"task","created_at":"2026-02-16T20:01:23.995556993-05:00","updated_at":"2026-02-16T20:02:19.933165582-05:00","closed_at":"2026-02-16T20:02:19.933165582-05:00","close_reason":"Closed"}
-{"id":"luminaguard-548","title":"Add rate limiting and quota management for agent operations","description":"Implement per-agent and per-user rate limiting with configurable quotas for API calls, VM spawning, and approval requests. Add admin dashboard to monitor and adjust quotas.","status":"open","priority":1,"issue_type":"task","estimated_minutes":240,"created_at":"2026-02-17T15:06:39.720517371-05:00","created_by":"alexc","updated_at":"2026-02-17T15:06:39.720517371-05:00","labels":["feature"]}
+{"id":"luminaguard-548","title":"Add rate limiting and quota management for agent operations","description":"Implement per-agent and per-user rate limiting with configurable quotas for API calls, VM spawning, and approval requests. Add admin dashboard to monitor and adjust quotas.","status":"in_progress","priority":1,"issue_type":"task","estimated_minutes":240,"created_at":"2026-02-17T15:06:39.720517371-05:00","created_by":"alexc","updated_at":"2026-02-19T13:48:33.580245817-05:00","labels":["feature"]}
 {"id":"luminaguard-5aj","title":"Implement Week 2: Security Code Execution Defense","description":"Validate LLM sanitization and tool execution security.\n\nTesting scope:\n- Inject malicious code into LLM prompts\n- Pass dangerous tool arguments\n- Attempt shell injection via MCP\n- Use fuzzing framework and prompt injection tester\n\nExpected: All malicious inputs blocked\n\nAcceptance criteria:\n- [ ] Fuzzing framework implemented\n- [ ] Prompt injection tests created\n- [ ] Shell injection tests created\n- [ ] All malicious inputs verified blocked\n- [ ] Security report generated\n- [ ] Results stored in .beads/metrics/security/\n\nRelated files: docs/validation/security-validation-plan.md","status":"closed","priority":1,"issue_type":"task","created_at":"2026-02-14T22:22:27.222124271-05:00","created_by":"alexc","updated_at":"2026-02-14T23:12:43.025902334-05:00","closed_at":"2026-02-14T23:12:43.025902334-05:00","close_reason":"Closed","dependencies":[{"issue_id":"luminaguard-5aj","depends_on_id":"luminaguard-ztm","type":"blocks","created_at":"2026-02-14T22:23:15.573142841-05:00","created_by":"alexc"}]}
 {"id":"luminaguard-5br","title":"Summary","description":"Create the main bot server entry point that runs all messenger connectors concurrently and manages the 24/7 operation of Luminaguard as an AI assistant bot.","status":"closed","priority":2,"issue_type":"task","created_at":"2026-02-16T20:01:45.141454569-05:00","updated_at":"2026-02-16T20:02:19.963152432-05:00","closed_at":"2026-02-16T20:02:19.963152432-05:00","close_reason":"Closed"}
 {"id":"luminaguard-5jc","title":"Implement Week 3: Reliability Network Partition Testing","description":"Verify network resilience under partition conditions.\n\nTesting scope:\n- Cut MCP connections during operation\n- Test agent behavior without connection\n- Verify fallback mechanisms\n- Test partial failure scenarios\n\nExpected: Graceful handling, no cascading failures\n\nAcceptance criteria:\n- [ ] Network partition test harness created\n- [ ] Connection loss handling verified\n- [ ] Fallback mechanisms tested\n- [ ] Partial failure scenarios tested\n- [ ] No cascading failures observed\n- [ ] Results stored in .beads/metrics/reliability/\n\nRelated files: docs/validation/reliability-testing-plan.md","status":"closed","priority":2,"issue_type":"task","created_at":"2026-02-14T22:22:28.075994569-05:00","created_by":"alexc","updated_at":"2026-02-14T23:12:43.036647483-05:00","closed_at":"2026-02-14T23:12:43.036647483-05:00","close_reason":"Closed","dependencies":[{"issue_id":"luminaguard-5jc","depends_on_id":"luminaguard-qj7","type":"blocks","created_at":"2026-02-14T22:23:16.319033552-05:00","created_by":"alexc"}]}

orchestrator/src/lib.rs

@@ -9,5 +9,6 @@ pub mod mcp;
 pub mod mcp_command;
 pub mod metrics;
 pub mod metrics_server;
+pub mod rate_limit;
 pub mod vm;
 pub mod webhooks;

orchestrator/src/rate_limit/config.rs

@@ -0,0 +1,201 @@
+//! Rate Limit Configuration
+//!
+//! Configuration for rate limiting and quota management.
+
+use serde::{Deserialize, Serialize};
+use std::time::Duration;
+
+/// Default rate limits
+pub const DEFAULT_API_RATE_LIMIT: u32 = 100; // requests per minute
+pub const DEFAULT_VM_SPAWN_LIMIT: u32 = 10; // VMs per hour
+pub const DEFAULT_APPROVAL_LIMIT: u32 = 50; // approvals per hour
+pub const DEFAULT_BURST_SIZE: u32 = 20; // burst allowance
+
+/// Rate limit configuration
+#[derive(Debug, Clone, Serialize, Deserialize)]
+pub struct RateLimitConfig {
+    /// Enable rate limiting
+    pub enabled: bool,
+
+    /// Default API rate limit (requests per minute)
+    pub default_api_limit: u32,
+
+    /// Default VM spawn limit (VMs per hour)
+    pub default_vm_spawn_limit: u32,
+
+    /// Default approval request limit (requests per hour)
+    pub default_approval_limit: u32,
+
+    /// Default burst size (allows temporary spikes)
+    pub default_burst_size: u32,
+
+    /// Token refill interval in seconds
+    pub refill_interval_secs: u64,
+
+    /// Quota reset interval in hours
+    pub quota_reset_interval_hours: u64,
+
+    /// Per-user quota overrides
+    #[serde(default)]
+    pub user_overrides: std::collections::HashMap<String, UserQuotaOverride>,
+
+    /// Per-agent quota overrides
+    #[serde(default)]
+    pub agent_overrides: std::collections::HashMap<String, AgentQuotaOverride>,
+}
+
+impl Default for RateLimitConfig {
+    fn default() -> Self {
+        Self {
+            enabled: true,
+            default_api_limit: DEFAULT_API_RATE_LIMIT,
+            default_vm_spawn_limit: DEFAULT_VM_SPAWN_LIMIT,
+            default_approval_limit: DEFAULT_APPROVAL_LIMIT,
+            default_burst_size: DEFAULT_BURST_SIZE,
+            refill_interval_secs: 60,
+            quota_reset_interval_hours: 24,
+            user_overrides: std::collections::HashMap::new(),
+            agent_overrides: std::collections::HashMap::new(),
+        }
+    }
+}
+
+impl RateLimitConfig {
+    /// Create a new rate limit configuration
+    pub fn new() -> Self {
+        Self::default()
+    }
+
+    /// Load configuration from environment variables
+    pub fn from_env() -> Self {
+        let mut config = Self::default();
+
+        if let Ok(val) = std::env::var("LUMINAGUARD_RATE_LIMIT_ENABLED") {
+            config.enabled = val.parse().unwrap_or(true);
+        }
+
+        if let Ok(val) = std::env::var("LUMINAGUARD_API_RATE_LIMIT") {
+            if let Ok(limit) = val.parse() {
+                config.default_api_limit = limit;
+            }
+        }
+
+        if let Ok(val) = std::env::var("LUMINAGUARD_VM_SPAWN_LIMIT") {
+            if let Ok(limit) = val.parse() {
+                config.default_vm_spawn_limit = limit;
+            }
+        }
+
+        if let Ok(val) = std::env::var("LUMINAGUARD_APPROVAL_LIMIT") {
+            if let Ok(limit) = val.parse() {
+                config.default_approval_limit = limit;
+            }
+        }
+
+        if let Ok(val) = std::env::var("LUMINAGUARD_BURST_SIZE") {
+            if let Ok(size) = val.parse() {
+                config.default_burst_size = size;
+            }
+        }
+
+        config
+    }
+
+    /// Get refill duration
+    pub fn refill_duration(&self) -> Duration {
+        Duration::from_secs(self.refill_interval_secs)
+    }
+
+    /// Get quota reset duration
+    pub fn quota_reset_duration(&self) -> Duration {
+        Duration::from_secs(self.quota_reset_interval_hours * 3600)
+    }
+
+    /// Disable rate limiting (for testing)
+    pub fn disabled() -> Self {
+        Self {
+            enabled: false,
+            ..Self::default()
+        }
+    }
+}
+
+/// Per-user quota override
+#[derive(Debug, Clone, Serialize, Deserialize)]
+pub struct UserQuotaOverride {
+    /// User ID
+    pub user_id: String,
+
+    /// API rate limit override
+    pub api_limit: Option<u32>,
+
+    /// VM spawn limit override
+    pub vm_spawn_limit: Option<u32>,
+
+    /// Approval limit override
+    pub approval_limit: Option<u32>,
+
+    /// Burst size override
+    pub burst_size: Option<u32>,
+
+    /// Is this user an admin (exempt from limits)
+    pub is_admin: bool,
+}
+
+/// Per-agent quota override
+#[derive(Debug, Clone, Serialize, Deserialize)]
+pub struct AgentQuotaOverride {
+    /// Agent ID
+    pub agent_id: String,
+
+    /// API rate limit override
+    pub api_limit: Option<u32>,
+
+    /// VM spawn limit override
+    pub vm_spawn_limit: Option<u32>,
+
+    /// Approval limit override
+    pub approval_limit: Option<u32>,
+
+    /// Burst size override
+    pub burst_size: Option<u32>,
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn test_default_config() {
+        let config = RateLimitConfig::default();
+        assert!(config.enabled);
+        assert_eq!(config.default_api_limit, DEFAULT_API_RATE_LIMIT);
+        assert_eq!(config.default_vm_spawn_limit, DEFAULT_VM_SPAWN_LIMIT);
+    }
+
+    #[test]
+    fn test_disabled_config() {
+        let config = RateLimitConfig::disabled();
+        assert!(!config.enabled);
+    }
+
+    #[test]
+    fn test_refill_duration() {
+        let config = RateLimitConfig::default();
+        assert_eq!(config.refill_duration(), Duration::from_secs(60));
+    }
+
+    #[test]
+    fn test_quota_reset_duration() {
+        let config = RateLimitConfig::default();
+        assert_eq!(config.quota_reset_duration(), Duration::from_secs(24 * 3600));
+    }
+
+    #[test]
+    fn test_config_serialization() {
+        let config = RateLimitConfig::default();
+        let json = serde_json::to_string(&config).unwrap();
+        let parsed: RateLimitConfig = serde_json::from_str(&json).unwrap();
+        assert_eq!(config.default_api_limit, parsed.default_api_limit);
+    }
+}