Listing Attacks

List All Attacks

tachyonic attacks

Shows all attacks available from your heuristics library.

Filter by Category

tachyonic attacks --category tool-abuse

Detailed View

tachyonic attacks --detailed

Shows full attack metadata: ID, name, category, severity, modality, payloads count, and OWASP mapping.

Include Built-in Attacks

By default, only the external heuristics library is loaded. Add built-in attacks:

tachyonic attacks --include-builtin

Example Output

prompt-injection
  PI-001  Direct Instruction Override         critical  text  4 payloads
  PI-002  System Prompt Extraction            high      text  3 payloads
  PI-003  Role Hijacking                      high      text  3 payloads
  ...

tool-abuse
  EA-001  Tool Parameter Manipulation         high      text  4 payloads
  EA-002  Tool Chaining Attack                critical  text  3 payloads
  ...

Categories

Library counts are for the default attack library. Built-in modules are added with --include-builtin. See the attack catalog for framework relations.

Category	Library	Built-in	Description
`prompt-injection`	20	3	Direct prompt injection
`system-prompt-extraction`	12	3	System prompt leakage
`jailbreak`	22	3	Guardrail bypass
`tool-abuse`	20	6	Tool/function manipulation
`multi-turn-manipulation`	10	2	Multi-turn attacks
`vision-injection`	12	0	Image-based attacks
`sensitive-disclosure`	10	4	Data exfiltration
`supply-chain`	12	0	Dependency attacks
`improper-output`	8	3	Output handling
`unbounded-consumption`	2	0	Resource exhaustion
`misinformation`	6	2	False information
`vector-embedding`	10	0	Embedding manipulation
`memory-context-poisoning`	6	0	Poisoned memory and context
`unexpected-code-execution`	6	0	Unintended code or command execution
`inter-agent-communication`	6	0	Attacks between cooperating agents
`human-trust-exploitation`	6	0	Output that manipulates the human reviewer
`permission-escalation`	0	6	Privilege escalation
`multi-agent-injection`	0	6	Multi-agent attacks
`indirect-injection`	0	4	Context injection (RAG)
`video-injection`	0	0	Reserved; no attacks yet
`transport-recon`	27	0	HTTP/MCP transport recon (opt-in, not in default scans)