Security

OWASP Agentic Top 10 Coverage

The OWASP Top 10 for Agentic Applications (2026) defines the critical security risks for AI agent systems. Below is how Daylite's modules map to each risk, with specific coverage status and the Rust modules responsible.

6
Fully Covered
3
Partial
1
Gap
IDRiskStatusDaylite Modules
ASI-01Agent Goal Hijacking⚠️ PartialScanner L1 (aho-corasick + regex), Shadow Scanner L2-L4, Inbound Response Scanner, MCP tool name validation
ASI-02Tool Misuse and Exploitation CoveredMCP Enforcement Proxy (5-step pipeline), Per-tool McpToolPolicy (allow/deny/audit), Tool list filtering, Rate Limiter (Redis Lua), Audit log
ASI-03Identity and Privilege Abuse CoveredRBAC + TenantContext, Phantom Token Vault, API Key Store (dyl_ prefix), SCIM 2.0, mTLS, OIDC SSO, Cell-based Tenancy
ASI-04Agentic Supply Chain Vulnerabilities CoveredMCP Server Registration (SSRF protection), Redirect SSRF blocking, Classification ceiling per server, Inbound Response Scanner, cargo-deny + cargo-audit, Docker cosign
ASI-05Unexpected Code Execution CoveredNo dynamic evaluation (architectural rule), Pipeline Executor (typed steps), NL Pipeline Generator (human gates), SecureBus, Template Engine (JSON-safe)
ASI-06Memory and Context Poisoning⚠️ PartialRAG Pipeline Security Scanning, PII Redactor (on ingest), Scanner (blocks secrets in documents), Classification Enforcement, Per-tenant KB isolation, BYOK Encryption
ASI-07Insecure Inter-Agent Communication CoveredSecureBus (compile-time enforcement), MCP Proxy (single gateway), HMAC-signed Audit Chain, mTLS, TLS (rustls + aws-lc-rs)
ASI-08Cascading Failures⚠️ PartialRate Limiter (Redis Lua token-bucket), Per-tool rate limits, Request timeouts, Backpressure (mpsc channels), Pipeline retry with backoff, 1MB DoS cap
ASI-09Human-Agent Trust Exploitation CoveredClassification Banners, Immutable Audit Trail, "AI Proposes, Human Gates" pattern, Webhook Alerts (SOC/SIEM), MCP audit-only mode
ASI-10Rogue Agents CoveredOutbound DLP Inspector, Echo-back Detection (Phantom Tokens), Immutable Audit Log (SHA-256 hash chain), SIEM Connector (Splunk HEC, OCSF), MITRE ATT&CK Events, Privacy-aware Routing

Detailed Coverage

ASI-01
Agent Goal Hijacking
⚠️ Partial
Scanner L1 (aho-corasick + regex), Shadow Scanner L2-L4, Inbound Response Scanner, MCP tool name validation

L1 pattern matching is production-ready (<1ms). L2-L4 ML layers (semantic injection detection, LlamaGuard content filter) are stubbed. L1 catches secrets/keys but not sophisticated semantic hijacking.

ASI-02
Tool Misuse and Exploitation
Covered
MCP Enforcement Proxy (5-step pipeline), Per-tool McpToolPolicy (allow/deny/audit), Tool list filtering, Rate Limiter (Redis Lua), Audit log

Full enforcement proxy with per-tool allow/deny/audit, argument inspection, classification enforcement, and rate limiting. Human-in-the-loop gating via require_approval field.

ASI-03
Identity and Privilege Abuse
Covered
RBAC + TenantContext, Phantom Token Vault, API Key Store (dyl_ prefix), SCIM 2.0, mTLS, OIDC SSO, Cell-based Tenancy

Phantom Token architecture is the key differentiator -- agents never hold real credentials. Per-tenant isolation + RBAC + SCIM instant deprovision addresses the full identity lifecycle.

ASI-04
Agentic Supply Chain Vulnerabilities
Covered
MCP Server Registration (SSRF protection), Redirect SSRF blocking, Classification ceiling per server, Inbound Response Scanner, cargo-deny + cargo-audit, Docker cosign

Full MCP server vetting (SSRF, classification ceiling, per-tool policy). Build-time supply chain hardened with cargo-deny/audit and cosign. Runtime tool responses scanned before reaching agent.

ASI-05
Unexpected Code Execution
Covered
No dynamic evaluation (architectural rule), Pipeline Executor (typed steps), NL Pipeline Generator (human gates), SecureBus, Template Engine (JSON-safe)

Rust memory safety + no dynamic eval + typed pipeline steps eliminate primary code execution vectors. Declarative connector template engine is JSON-escaped by design. Human approval gate on generated pipelines.

ASI-06
Memory and Context Poisoning
⚠️ Partial
RAG Pipeline Security Scanning, PII Redactor (on ingest), Scanner (blocks secrets in documents), Classification Enforcement, Per-tenant KB isolation, BYOK Encryption

Document-level scanning (PII + secrets + classification) on ingest is implemented. Missing: adversarial embedding detection, document provenance tracking, periodic re-scanning of existing KB content.

ASI-07
Insecure Inter-Agent Communication
Covered
SecureBus (compile-time enforcement), MCP Proxy (single gateway), HMAC-signed Audit Chain, mTLS, TLS (rustls + aws-lc-rs)

SecureBus provides compile-time enforcement unique in the market -- Rust type system makes compliance bypass impossible. Monolith architecture means inter-agent communication is in-process (no network attack surface).

ASI-08
Cascading Failures
⚠️ Partial
Rate Limiter (Redis Lua token-bucket), Per-tool rate limits, Request timeouts, Backpressure (mpsc channels), Pipeline retry with backoff, 1MB DoS cap

Rate limiting, timeouts, backpressure, and DoS caps are implemented. Missing: automatic circuit breaker (currently manual enabled flag), cascading failure detection across multi-step pipelines, formal chaos testing.

ASI-09
Human-Agent Trust Exploitation
Covered
Classification Banners, Immutable Audit Trail, "AI Proposes, Human Gates" pattern, Webhook Alerts (SOC/SIEM), MCP audit-only mode

The "AI Proposes, Human Gates" pattern directly addresses trust exploitation by requiring human approval for consequential actions. Classification banners and audit trails provide transparency.

ASI-10
Rogue Agents
Covered
Outbound DLP Inspector, Echo-back Detection (Phantom Tokens), Immutable Audit Log (SHA-256 hash chain), SIEM Connector (Splunk HEC, OCSF), MITRE ATT&CK Events, Privacy-aware Routing

Multi-layer exfiltration defense: outbound DLP + phantom token echo-back detection + classification-enforced routing + immutable audit. MITRE ATT&CK mapping enables SOC correlation.

Gap Roadmap

ASI-01: Scanner L2-L4 (ML Injection Detection)
Q2 2026
Implement embedding similarity (L2), DistilBERT classifier (L3), and LlamaGuard content filter (L4) to move from regex-only to ML-based prompt injection detection.
ASI-06: Adversarial Embedding Detection
Q3 2026
Add vector anomaly detection, document provenance tracking, and periodic re-scan jobs for existing knowledge base content.
ASI-08: Automatic Circuit Breaker
Q2 2026
Implement automatic circuit breaker that trips on error rate threshold rather than requiring manual toggle. Add pipeline-level health scoring.

Competitive Differentiators

Phantom Tokens (ASI-03)
Agents never hold real credentials. No competitor offers this at the proxy layer.
SecureBus (ASI-07)
Rust type system makes compliance bypass impossible. Cannot be replicated in Python/Go/Java.
MCP Enforcement Proxy (ASI-02)
First-mover on MCP security with full 5-step inspection pipeline.
Single Binary (ASI-08)
No inter-service network to fail. Eliminates an entire class of cascading failure modes.
Trust CenterBack to Daylite