SIREN

Your agent reads the web. SIREN controls what it finds. Indirect prompt injection. Weaponised at scale.
5
Target Actions
5
Subsystems
8
Hiding Techniques
58
Tests Passing
pip install red-specter-siren
AI agents browse the web / Read documents you don't control / Process external content you haven't inspected / Indirect injection hides in plain sight / White-on-white text. CSS visibility zero. HTML comments / The payload is already on the page / Your agent will read it / SIREN wrote it AI agents browse the web / Read documents you don't control / Process external content you haven't inspected / Indirect injection hides in plain sight / White-on-white text. CSS visibility zero. HTML comments / The payload is already on the page / Your agent will read it / SIREN wrote it

Your Agent Reads Content You Don't Control

AI agents browse the web, read documents, process email, and consume external data sources. None of that content is trusted. Any of it can contain instructions directed at the agent. The agent cannot tell the difference between legitimate content and malicious instructions hidden inside it. SIREN tests whether your agent can be made to do something it should not do — triggered entirely by content it retrieved from an external source.

The Agent Cannot See the Injection

Indirect injection hides in white-on-white text, zero-opacity CSS, HTML comments, metadata fields, image alt text, and zero-width Unicode characters. The user sees a normal webpage. The agent sees the instructions hidden inside it.

External Content Is Untrusted by Default

Web pages, PDF documents, emails, API responses, database records — every external source is a potential injection vector. Agents that browse, read, and act are permanently exposed to every page they visit.

Five Actions That Cause Real Damage

Data exfiltration. Credential harvesting. Privilege escalation. Lateral movement. Persistent backdoor installation. SIREN tests all five. Each one represents a class of attack that an agent could execute entirely without the user's knowledge.

No Existing Tool Tests This Surface

Direct prompt injection tools exist. Indirect injection weaponisation — the creation and delivery of payloads embedded in external content retrieved by agents — has no dedicated offensive tooling. SIREN is the first.

External Content as a Weapon

SIREN operates across five subsystems. CRAFTER builds the malicious content. HIDER embeds payloads using eight concealment techniques. COURIER delivers them to the agent's retrieval path. TRIGGER activates them at the right moment. EVIDENCE documents everything for the signed report.

01

CRAFTER

PAYLOAD CONSTRUCTION

Builds indirect injection payloads for five target action classes: data exfiltration, credential harvesting, privilege escalation, lateral movement, and persistent backdoor installation. Every payload is crafted for the specific agent architecture under test.

02

HIDER

8 CONCEALMENT TECHNIQUES

White-on-white text. CSS visibility zero. HTML comments. Metadata embedding. Image alt text injection. Zero-width Unicode characters. Markdown hidden syntax. Base64 in data attributes. Eight ways to hide an instruction in plain sight.

03

COURIER

DELIVERY ENGINE

Positions malicious content in the agent's retrieval path. Web pages, documents, email bodies, API response payloads, database records. COURIER ensures the agent encounters the injection through its normal retrieval behaviour — not through a special attack channel.

04

TRIGGER

ACTIVATION ENGINE

Times payload activation to maximise impact. Context-aware triggers fire when the agent has sufficient privileges, an active session, or a relevant task in progress. Dormant injections that activate on specific agent state are TRIGGER's speciality.

05

EVIDENCE

REPORT & SIGN

Captures full evidence of every successful injection. Agent behaviour before and after. Payload delivery confirmation. Action executed. Ed25519 signed report. OWASP LLM01 mapping. SHA-256 evidence chain for every finding.

ARMORY

NIGHTFALL ARMORY Integration

970 PAYLOADS — CENTRALISED LIBRARY

Connected to the NIGHTFALL ARMORY payload intelligence library. SIREN pulls indirect injection and content poisoning payloads on demand. Successful injections feed back into ARMORY for fleet-wide improvement. Every SIREN finding strengthens every tool in NIGHTFALL.

Craft. Hide. Deliver. Trigger. Report.

SIREN runs a complete indirect injection engagement. CRAFTER builds payloads for the target action. HIDER conceals them. COURIER places them in the agent's path. TRIGGER activates. EVIDENCE documents and signs.

Full indirect injection engagement against an agent with web browsing capability:

$ siren craft --action exfiltrate --target-agent https://agent.example.com
$ siren hide --technique css-zero-opacity --payload payload.json
$ siren deliver --courier web --trigger context-aware --override
[CRAFTER] Building exfiltration payload...
  Target action: data exfiltration | 5 payload variants generated
[HIDER] Concealing with CSS zero-opacity technique...
  Payload embedded — invisible to human reader, visible to LLM tokeniser
[COURIER] Placing payload in agent retrieval path...
  Delivery channel: web page — agent will retrieve on next browse
[TRIGGER] Context-aware trigger armed...
  INJECTION ACTIVATED — agent retrieved and processed malicious content
  Exfiltration confirmed — 847 bytes extracted to attacker endpoint
[EVIDENCE] Capturing evidence chain...

ENGAGEMENT COMPLETE | OWASP LLM01 | Report signed ✓ | SHA-256 chain verified
5
Target Actions
5
Subsystems
8
Hiding Techniques
58
Tests Passing
0
Failures

UNLEASHED Clearance

Every injection attempt is Ed25519 signed, scope-locked to authorised agents, and auto-locks after 30 minutes. Three tiers of operation. Authorised penetration testing only. EVIDENCE is mandatory — SIREN logs every payload and every outcome.

Detection

Scans agent architecture for indirect injection susceptibility. Maps retrieval paths. No payload delivery. Full surface report in a signed document.

Dry Run

Crafts and hides payloads. Shows exact delivery path and trigger conditions. Ed25519 required. No injection executed. Full projected impact shown.

Live Execution

Full indirect injection engagement. COURIER delivers. TRIGGER activates. EVIDENCE logs. RESTRICTED signed report with all successful injections documented.

THIS TOOL IS FOR AUTHORISED SECURITY TESTING ONLY. EVERY EXECUTION IS SIGNED AND LOGGED.

Ed25519 Cryptographic Override
SIREN UNLEASHED

Cryptographic override. Private key controlled. One operator. Founder's machine only.

Security Distros & Package Managers

Kali Linux
.deb package
Parrot OS
.deb package
BlackArch
PKGBUILD
REMnux
.deb package
PyPI
pip install
macOS
pip install
Windows
pip install
Docker
docker pull

Authorised Use Only

Red Specter SIREN is intended for authorised security testing only. Deploying indirect prompt injection payloads against AI agents or systems you do not own or have explicit written 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. Apache License 2.0.