ECHO

RAG poisoning attack engine — corrupt the memory layer. Your AI remembers everything. ECHO decides what.
7
Subsystems
36
Techniques
211
Tests
31
Tool in Suite
pip install red-specter-echo
Your vector database has no integrity checks / Poisoned embeddings surface in every retrieval / Context windows trust whatever they retrieve / RAG pipelines have no tamper detection / Persistent memory is permanently writable / Knowledge base ingestion is unauthenticated / Your AI remembers everything. ECHO decides what. Your vector database has no integrity checks / Poisoned embeddings surface in every retrieval / Context windows trust whatever they retrieve / RAG pipelines have no tamper detection / Persistent memory is permanently writable / Knowledge base ingestion is unauthenticated / Your AI remembers everything. ECHO decides what.

RAG Pipelines Have No Integrity Layer

Every organisation deploying RAG-augmented AI assumes the vector database is trustworthy. It isn't. Documents enter the index without authentication. Embeddings are stored without integrity checks. Retrieval results are trusted implicitly. ECHO weaponises every assumption in the RAG pipeline.

Unauthenticated Vector Stores

Vector databases accept document ingestion without source authentication. An attacker with write access — or indirect access via an injection point — can plant any content into the semantic index. Retrieval will surface it ranked above legitimate documents.

Embedding Space Is Malleable

Semantic search relies on cosine similarity in high-dimensional embedding space. Adversarially crafted embeddings can occupy the same neighbourhood as legitimate content, crowding out genuine results and steering retrieval toward attacker-controlled content.

Context Windows Trust Retrieval

When retrieved content lands in the context window, the LLM treats it as trusted input. Poisoned retrieved documents carry instructions that override system prompts. The model follows the retrieved instruction, not its configured behaviour.

Persistent Memory Has No Rollback

Conversation history, long-term memory, and knowledge consolidation systems write to persistent stores without versioning. Injected content in persistent memory survives session boundaries and affects every subsequent interaction with the system.

Knowledge Base Ingestion Is Blind

Batch ingestion pipelines process documents without content validation. Malicious documents masquerading as legitimate sources pass through indexing intact. Once indexed, they cannot be distinguished from genuine content without semantic auditing.

Re-ranking Amplifies the Attack

Modern RAG pipelines use re-ranking models to score retrieved chunks. ECHO's RETRIEVE subsystem games relevance scores, exploits chunk boundary positioning, and spoofs source authority to ensure poisoned content ranks at the top of every retrieval.

Corrupt the Memory, Control the Model

ECHO targets the memory layer that RAG-augmented AI systems depend on. Poison the vector database and every retrieval returns your content. Manipulate embeddings and semantic search serves your payload. Hijack the context window and the model follows your instructions, not its own.

01

VECTOR

Vector DB Attacks

Inject malicious documents into vector stores. Similarity poisoning. Nearest-neighbour manipulation. Index corruption. Metadata tampering. Namespace collision attacks across multi-tenant vector databases.

02

EMBED

Embedding Manipulation

Adversarial embedding generation. Semantic space pollution. Cosine similarity exploitation. Dimension collapse attacks that degrade search quality. Embedding inversion to reconstruct original training data.

03

RETRIEVE

Retrieval Poisoning

Query manipulation to surface poisoned content. Relevance score gaming via adversarial chunk construction. Chunk boundary exploitation. Re-ranking attacks. Source authority spoofing to elevate attacker content.

04

CONTEXT

Context Window Hijacking

Context overflow attacks. Priority injection via retrieved content. Instruction smuggling embedded in retrieved documents. Attention steering. System prompt dilution via high-volume benign content injection.

05

PERSIST

Memory Corruption

Persistent memory poisoning via crafted conversation turns. Conversation history manipulation. Long-term memory injection that survives session boundaries. Memory consolidation attacks. Forgetting induction via targeted interference.

06

INJECT

Knowledge Base Injection

Knowledge base poisoning via batch ingestion exploitation. Document injection via trusted source impersonation. Update pipeline attacks targeting incremental indexing. Content authority spoofing to elevate injected documents.

07

ANTIDOTE

Mandatory Restore

Baseline vector store snapshot before any engagement. Embedding integrity verification pre and post attack. Retrieval quality measurement across all attack vectors. Signed restoration certificate with Ed25519.

Every Layer of the RAG Stack

ECHO maps 36 discrete attack techniques across the full RAG pipeline — from document ingestion through embedding, indexing, retrieval, re-ranking, and context assembly. Each technique is independently testable and maps to a specific defensive control.

# Technique Subsystem Attack Surface
01 Index Injection echo vector inject Direct document insertion into vector index bypassing ingestion authentication. Targets unauthenticated write endpoints on Chroma, Weaviate, Pinecone, Qdrant, and pgvector.
02 Similarity Poisoning echo embed poison Crafts adversarial documents whose embeddings land in the same semantic neighbourhood as target queries. Guarantees retrieval without matching keywords — purely semantic camouflage.
03 Chunk Boundary Exploit echo retrieve chunk Malicious instructions placed at chunk boundaries where splitting algorithms predictably divide documents. Ensures instructions land at the start of retrieved chunks for maximum LLM attention weight.
04 Context Overflow echo context overflow Floods the context window with high-volume benign content to dilute system prompt instructions. Attacker payload is positioned at optimal attention positions within the diluted context.
05 Memory Consolidation Attack echo persist consolidate Exploits memory consolidation timing in systems like MemGPT. Injects false memories during the consolidation window that are permanently merged into long-term storage.
06 Authority Spoofing echo inject authority Fabricates document metadata to impersonate high-trust sources. Exploits re-ranking models that weight source authority. Injected documents rank as internal documentation from authoritative sources.
07 Embedding Inversion echo embed invert Recovers approximate original text from stored embeddings using gradient-based inversion. Maps the reconstruction fidelity of the embedding model to assess data leakage risk from the vector store.
7
Subsystems
36
Techniques
211
Tests
6+
Vector DB Targets
0
Failures

UNLEASHED Gate

Standard mode maps RAG attack surfaces without exploitation. UNLEASHED executes poisoning campaigns. Ed25519 cryptographic gate. Dual-gate safety system. One operator. ANTIDOTE subsystem restores baseline vector state after every live engagement.

Detection

Maps RAG attack surfaces. Identifies vulnerable vector stores, unauthenticated ingestion pipelines, and retrieval systems. Measures embedding integrity. No exploitation — reports only.

Dry Run

Plans full poisoning campaigns. Shows exactly which injection points work, which queries surface poisoned content, and which memory corruption techniques persist. Ed25519 required. No execution.

Live Execution

Cryptographic override. Private key controlled. One operator. Founder's machine only. ANTIDOTE automatically captures baseline and restores vector state after live engagement completes.

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

Every Finding Mapped

OWASP LLM Top 10

AI Security Mapping

  • LLM01 Prompt Injection via retrieval
  • LLM02 Sensitive Information Disclosure
  • LLM04 Data and Model Poisoning
  • LLM05 Improper Output Handling
  • LLM07 System Prompt Leakage
  • LLM08 Vector and Embedding Weaknesses
MITRE ATLAS

Adversarial ML Coverage

  • AML.T0043 Craft Adversarial Data
  • AML.T0051 LLM Prompt Injection
  • AML.T0048 AI System Compromise
  • AML.T0020 Poison Training Data
  • AML.T0057 ML Artifact Collection
  • AML.T0040 Network-based Exfiltration
Cryptographic

Report Integrity

  • Ed25519 digital signatures
  • SHA-256 evidence chains
  • RFC 3161 timestamps
  • Tamper-evident by design
  • AI Shield policy generation
  • Machine-ingestible JSON output

Security Distros & Package Managers

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

Authorised Use Only

Red Specter ECHO is intended for authorised security testing only. Unauthorised use against 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. Apache License 2.0.

Ed25519 Cryptographic Override
ECHO UNLEASHED

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

7
Subsystems
36
Techniques
211
Tests Passing
0
External Dependencies