T-3001ObservedDeclining

System Prompt Credential Extraction

Extract credentials hardcoded or referenced in the agent's system prompt

Tactic

Credential Harvest · Stage 3

Extract API keys, tokens, and credentials from agent context and connected services

Attack class

RETROACTIVE-PRIV

Exploiting previously granted access or cached credentials to gain unauthorized capabilities

Evidence grade
Observed

Confirmed in real-world production systems or internet-wide exposure assessments.

DVAA validation

L1-02

Reproductions in Damn Vulnerable AI Agent, the OpenA2A intentionally-broken agent for kill-chain validation.

Honeypot

AgentPwn coverage

Live
data-exfiltrationagentpwn.com/learn ↗

An AgentPwn trap page produces a payload tagged with this technique class. Following the AgentPwn taxonomy of trap pages shows what an agent encounters.

Credential-probe tiers extract secrets referenced in the system prompt.

Detect

Detection · HackMyAgent

Live4 live · 0 queued
CRED-001CRED-002CRED-003CRED-004
npx hackmyagent secure --ciLive = implemented in hackmyagent; queued = declared
Defend

Defense · OASB controls

Live5 live · 0 queued
Live = documented at oasb.ai; queued = declared
Reference

How to cite

AI Agent Threat Matrix T-3001 (System Prompt Credential Extraction). OpenA2A, 2026. https://threats.opena2a.org/techniques/T-3001