GLASS

See everything. Intercept anything. Attack through the wire.
7
Protocols
850
Tests
10
Scanners
13
Models Tracked
pip install red-specter-glass
View Docs
Agent traffic is invisible / You can't intercept LLM API calls / MCP messages fly blind / Agent-to-agent traffic is unaudited / Credentials leak in plaintext / System prompts ship in every request / Tool calls execute without inspection / Nobody sits in the middle Agent traffic is invisible / You can't intercept LLM API calls / MCP messages fly blind / Agent-to-agent traffic is unaudited / Credentials leak in plaintext / System prompts ship in every request / Tool calls execute without inspection / Nobody sits in the middle

Nobody Sits in the Middle

Every AI security tool tests from the outside. Run a scan. Get a report. But nobody watches the wire in real time. Your AI agents are making hundreds of API calls per minute — to LLM providers, MCP servers, tool endpoints, and other agents — and you have no visibility into any of it. You are deploying agents whose traffic you have never seen.

Invisible Agent Traffic

Your agent talks to GPT-4, Claude, Gemini, local models, MCP servers, and other agents. You have never inspected a single message. You don't know what your agent is sending or receiving.

Credentials in the Clear

API keys, bearer tokens, session credentials — flowing through every request. No interception. No scanning. No alerts. You would never run a web app without a proxy. Why are you running AI agents without one?

System Prompts Everywhere

Your carefully crafted system prompt — with all your business logic, guardrails, and proprietary instructions — is sent in plaintext with every API call. And you have never looked at what comes back.

What Agents Really Talk To

Your agents make calls you never authorised. Here is what flies across the wire, unchecked:

  • LLM providers (OpenAI, Anthropic, Google)
  • MCP servers (tool registries, action endpoints)
  • External tools (APIs, databases, file systems)
  • Other agents (A2A protocol, swarm coordination)

The Proxy Position

GLASS sits between your AI agent and everything it talks to. Every message passes through GLASS — decoded, inspected, scanned, logged. Pause it. Modify it. Drop it. Replay it. The wire is yours.

AI Agent
GLASS Proxy
LLM / MCP / Tools / Agents
GLASS Dashboard
Real-time traffic view + alerts + session recording

Protocol-Native Decoding

GLASS does not just proxy HTTP. It decodes AI protocols natively — reassembling streaming responses, parsing JSON-RPC messages, and understanding the structure of every conversation. Not packet capture. Protocol intelligence.

OpenAI Chat Completions

Full decode of chat completion requests and responses. Streaming SSE chunk reassembly. Function calls, tool use, and system prompts extracted and displayed in real time.

streaming + non-streaming

Anthropic Messages API

Native decode of Anthropic's Messages API. System prompts, tool use blocks, and content blocks parsed individually. Streaming event reconstruction.

protocol-native decoding

Google Gemini API

Gemini generateContent and streamGenerateContent decoded natively. Safety ratings, citation metadata, and grounding results extracted per response.

protocol-native decoding

MCP JSON-RPC 2.0

Model Context Protocol messages decoded at the JSON-RPC layer. Tool registrations, capability negotiations, and action invocations fully parsed and inspectable.

protocol-native decoding

LangChain / LangGraph

LangChain callback traffic and LangGraph state transitions intercepted and decoded. Chain step visibility, agent reasoning traces, and tool call sequences.

protocol-native decoding

A2A (Agent-to-Agent)

Google's Agent-to-Agent protocol decoded natively. Task lifecycle, artifact exchange, and capability discovery messages fully visible and modifiable.

protocol-native decoding

Generic HTTP/HTTPS

Any HTTP/HTTPS traffic proxied with full TLS interception. Request/response pairs captured, headers inspected, bodies decoded. The baseline for everything else.

universal fallback
7
AI Protocols
10
Passive Scanners
850
Tests Passing
13
Models Tracked
0
Equivalents

Intercept. Inspect. Modify. Forward.

GLASS gives you the same power over AI agent traffic that Burp Suite gives you over web traffic. Pause any message. Read it. Change it. Send it. Or drop it entirely. Full control over the wire.

01

Live Intercept & Modify

Pause any message in transit — request or response. Inspect the full decoded payload. Modify any field: system prompts, user messages, tool calls, function arguments. Forward the modified version or drop it entirely. Man-in-the-middle for AI agents.

02

Session Recording & Replay

Capture entire agent sessions — every request, every response, every tool call, every streaming chunk. Replay sessions with modifications. Inject payloads into recorded sessions. Build regression test suites from real traffic.

03

Match & Replace Rules

Automated traffic modification. Define rules that match patterns in requests or responses and replace them on the fly. Swap system prompts. Inject instructions. Modify tool responses. Runs continuously without manual intervention.

04

Streaming Decode

SSE (Server-Sent Events) chunk reassembly in real time. See the complete response as it builds, not as a stream of fragments. Works across OpenAI, Anthropic, and Gemini streaming endpoints. Protocol-native, not regex.

Zero-Config. Every Message Scanned.

Every message that passes through GLASS is automatically scanned by 10 passive detection engines. No configuration required. No rules to write. Alerts fire in real time on the dashboard. Findings are recorded in the session log with full evidence.

Credential Leak

API keys, bearer tokens, passwords, and secrets detected in request and response bodies.

10 patterns

PII Detection

Names, emails, phone numbers, addresses, SSNs, and other personal data flagged in transit.

8 patterns

System Prompt Exposure

Detects when system prompts are leaked in model responses — partial or full extraction.

Prompt Injection

Known injection patterns detected in incoming messages and tool responses before they reach the agent.

Tool Abuse

Detects suspicious tool invocation patterns — privilege escalation, file system access, network calls to unexpected endpoints.

Data Exfiltration

Detects data being extracted through model responses, tool outputs, or agent-to-agent messages.

Anomaly Detection

Baseline traffic patterns established per session. Deviations in message size, frequency, and content flagged automatically.

Token Cost Tracking

Real-time token counting and cost estimation across all proxied traffic. Per-session, per-model, per-agent breakdowns.

13 models

Sensitive Content

Toxicity, bias, and harmful content detected in both agent outputs and model responses using pattern matching.

Model Fingerprinting

Detects model substitution by fingerprinting response patterns. Alerts if the model behind the API changes unexpectedly.

Ten Tools. Every Layer. Nothing Assumed Safe.

GLASS is not a stage. It is the lens through which every stage can be focused. Point FORGE through GLASS to see every payload in transit. Route ARSENAL through GLASS to watch agent exploitation in real time. GLASS amplifies every tool in the pipeline.

Stage 1 — LLM Testing
FORGE
LLM Testing
Stage 2 — Agent Testing
ARSENAL
Agent Testing
Stage 3 — Swarm Assault
PHANTOM Swarm
Swarm Assault
Stage 4 — Web Siege
POLTERGEIST
Web Siege
Stage 5 — Traffic Interception
GLASS
Traffic Interception
Stage 6 — Adversarial AI
NEMESIS
Adversarial AI
Stage 7 — Human Layer
SPECTER SOCIAL
Human Layer
Stage 8 — OS/Kernel
PHANTOM KILL
OS/Kernel
Stage 9 — Physical Layer
GOLEM
Physical Layer
Stage 10 — Supply Chain
HYDRA
Supply Chain
IDRIS
IDRIS
Discovery & Governance
Defence
AI Shield
Defence
SIEM
redspecter-siem
SIEM Integration
GLASS Position
GLASS sits across every stage. Route any tool through the proxy to gain full wire-level visibility. See what FORGE sends. See what ARSENAL exploits. See what AI Shield blocks. See what reaches production.

Real-Time Traffic View

Every proxied message displayed in real time. Protocol-decoded. Colour-coded by status. Alerts fire on the right panel when scanners detect issues. Click any row to inspect the full decoded message.

1,247
Messages
3
Alerts
42.8K
Tokens
$1.24
Cost
RECORDING
Session
# Protocol Host Method Status Size Time
1247 OpenAI api.openai.com POST /v1/chat/completions 200 4.2 KB 142ms
1246 MCP localhost:3001 JSON-RPC tools/call 200 892 B 23ms
1245 Anthropic api.anthropic.com POST /v1/messages 200 6.8 KB 890ms
1244 A2A agent-beta:8080 POST /tasks/send 200 1.1 KB 45ms
1243 OpenAI api.openai.com POST /v1/chat/completions 200 3.9 KB 1.2s
1242 MCP localhost:3001 JSON-RPC tools/list 200 2.4 KB 12ms
1241 Gemini generativelanguage.googleapis.com POST /v1/models/gemini-pro:generate 200 5.1 KB 670ms
Alerts (3)
Credential Leak
#1243 — API key found in response body
PII Detected
#1245 — Email address in user message
System Prompt
#1245 — Partial system prompt in response

Five Commands. Full Control.

Everything through the CLI. Start a proxy. Intercept a session. Replay with injected payloads. Scan recorded traffic. Generate a signed report. Pipe it into your SIEM.

# Start proxy with protocol-native decoding
$ glass proxy --port 8080 --protocols openai,mcp,a2a

# Live intercept with modification
$ glass intercept --session pentest_01 --modify

# Replay session with FORGE payloads injected
$ glass replay --session pentest_01 --inject forge_payloads.yaml

# Scan recorded session with ARSENAL
$ glass scan --session pentest_01 --tool arsenal

# Generate signed report and export to SIEM
$ glass report --session pentest_01 --sign --export-siem splunk

Cryptographic Certainty

Every session recorded by GLASS is cryptographically signed. Every finding is timestamped. Every alert is mapped to MITRE ATLAS. Every report exports natively to your SIEM. Evidence that holds up under audit.

Ed25519 Signed Sessions

Every recorded session cryptographically signed with Ed25519. Tamper-evident by design. Verify that no message was added, removed, or modified after capture.

RFC 3161 Timestamps

Trusted timestamping for every session and finding. Prove when traffic was captured. Prove when alerts fired. Immutable temporal evidence chain.

MITRE ATLAS Mapping

Every scanner finding mapped to MITRE ATLAS techniques. Machine-learning attack taxonomy applied automatically. Consistent with industry-standard threat classification.

SIEM Export

Native export to Splunk (HEC/CIM), Microsoft Sentinel (CEF/Log Analytics), and IBM QRadar (LEEF 2.0/Syslog). One flag. Signatures preserved across export.

Five Reasons This Did Not Exist

01

Protocol-Native

Decodes AI protocols, not just HTTP. Understands conversations, not packets.

02

First Mover

No equivalent exists. There is no Burp Suite for AI agents. Until now.

03

Stack Integration

Amplifies FORGE, ARSENAL, PHANTOM, and POLTERGEIST. Every tool gains wire visibility.

04

Evidence-Grade

Ed25519 + RFC 3161. Sessions that hold up under legal and compliance audit.

05

Burp Analogy

Instantly understood by every security practitioner on earth. No explanation needed.

Security Distros & Package Managers

Kali Linux
.deb package
Parrot OS
.deb package
BlackArch
PKGBUILD
REMnux
.deb package
Tsurugi
.deb package
PyPI
pip install

Authorised Use Only

Red Specter GLASS is intended for authorised security testing and traffic inspection only. Intercepting network traffic on systems you do not own or have explicit permission to test may violate the Computer Misuse Act 1990 (UK), Computer Fraud and Abuse Act (US), and equivalent legislation in other jurisdictions. Always obtain written authorisation before conducting any security assessments or traffic interception. Apache License 2.0.

Ready to See What Your Agents Are Really Saying?

Every AI agent has traffic you have never inspected. Every LLM call carries data you have never seen. GLASS gives you the wire. See everything. Intercept anything. Attack through the wire.

Contact