OWASP ASI Classification Framework
OWASP Top 10 for Agentic Applications 2026 - Standardized risk classification for AI agent security.
Risk Categories
| ID |
Risk Type |
Key Indicators |
| ASI01 |
Agent Goal Hijack |
Prompt injection, instruction override, goal manipulation |
| ASI02 |
Tool Misuse & Exploitation |
Unauthorized tool calls, parameter tampering, unvalidated inputs |
| ASI03 |
Identity & Privilege Abuse |
Auth bypass, permission escalation, missing authorization |
| ASI04 |
Agentic Supply Chain |
Malicious dependencies, compromised tools, package poisoning |
| ASI05 |
Unexpected Code Execution |
RCE, command injection, code evaluation |
| ASI06 |
Memory & Context Poisoning |
Data leakage, context manipulation, memory corruption |
| ASI07 |
Insecure Inter-Agent Comm |
Unencrypted channels, data exposure between agents |
| ASI08 |
Cascading Failures |
Error propagation, chain reaction vulnerabilities |
| ASI09 |
Human-Agent Trust Exploit |
Social engineering, deceptive responses |
| ASI10 |
Rogue Agents |
Malicious agent behavior, unauthorized actions |
Detection Source → ASI Mapping
| Detection Source |
Type |
Primary ASI |
Secondary ASI |
data-leakage-detection |
Skill |
ASI06, ASI07 |
ASI01, ASI03 |
tool-abuse-detection |
Skill |
ASI02, ASI05, ASI07 |
ASI03 |
indirect-injection-detection |
Skill |
ASI01 |
ASI06 |
authorization-bypass-detection |
Skill |
ASI03 |
ASI09 |
agentic-supply-chain-detection |
Skill |
ASI04 |
ASI10 |
unexpected-code-execution-detection |
Skill |
ASI05 |
ASI04 |
inter-agent-comm-security-detection |
Skill |
ASI07 |
ASI06 |
cascading-failure-detection |
Skill |
ASI08 |
ASI10 |
human-agent-trust-exploit-detection |
Skill |
ASI09 |
ASI01 |
| Prompt Injection tests |
Dialogue |
ASI01, ASI06 |
ASI09 |
| Code Audit |
Agent |
ASI04, ASI05 |
ASI10 |
Finding → ASI Mapping
| Finding Type |
ASI Category |
Rationale |
| API keys, tokens |
ASI06 |
Context contains sensitive data |
| System prompts |
ASI01 |
Enables goal hijacking |
| Credentials |
ASI03 |
Identity abuse risk |
| Internal configs |
ASI04 |
Supply chain exposure |
| PII exposure |
ASI07 |
Inter-agent data leak |
| Command injection |
ASI05 |
Unexpected code execution |
| Unauthorized tool calls |
ASI02 |
Tool misuse |
Severity Classification
Critical 🚨
- Complete private keys exposed
- Production credentials with unrestricted access
- Remote code execution confirmed
- Full agent takeover possible
Action: Immediate remediation (within 24 hours)
High ⚠️
- API keys with limited scope
- Authentication tokens
- System prompt disclosure
- Partial credential exposure
Action: Urgent remediation (within 1 week)
Medium 💡
- Configuration details
- Internal endpoints
- Metadata leakage
- Conditional exploitation paths
Action: Address within 2-4 weeks
Low ℹ️
- Generic system information
- Non-sensitive configurations
- Indirect hints
Action: Review as time permits
Usage
Load this skill when performing OWASP ASI classification:
load_skill(name="owasp-asi")
Then apply the mapping rules to classify findings.
1---2name: owasp-asi3description: OWASP Top 10 for Agentic Applications 2026 (ASI) classification framework. Use for mapping security findings to standardized risk categories.4metadata:5 source: https://skillmd.com/skills/tencent-ai-infra-guard/owasp-asi6---78# OWASP ASI Classification Framework910**OWASP Top 10 for Agentic Applications 2026** - Standardized risk classification for AI agent security.1112## Risk Categories1314| ID | Risk Type | Key Indicators |15|:---|:----------|:---------------|16| **ASI01** | Agent Goal Hijack | Prompt injection, instruction override, goal manipulation |17| **ASI02** | Tool Misuse & Exploitation | Unauthorized tool calls, parameter tampering, unvalidated inputs |18| **ASI03** | Identity & Privilege Abuse | Auth bypass, permission escalation, missing authorization |19| **ASI04** | Agentic Supply Chain | Malicious dependencies, compromised tools, package poisoning |20| **ASI05** | Unexpected Code Execution | RCE, command injection, code evaluation |21| **ASI06** | Memory & Context Poisoning | Data leakage, context manipulation, memory corruption |22| **ASI07** | Insecure Inter-Agent Comm | Unencrypted channels, data exposure between agents |23| **ASI08** | Cascading Failures | Error propagation, chain reaction vulnerabilities |24| **ASI09** | Human-Agent Trust Exploit | Social engineering, deceptive responses |25| **ASI10** | Rogue Agents | Malicious agent behavior, unauthorized actions |2627## Detection Source → ASI Mapping2829| Detection Source | Type | Primary ASI | Secondary ASI |30|:-----------------|:-----|:------------|:--------------|31| `data-leakage-detection` | Skill | ASI06, ASI07 | ASI01, ASI03 |32| `tool-abuse-detection` | Skill | ASI02, ASI05, ASI07 | ASI03 |33| `indirect-injection-detection` | Skill | ASI01 | ASI06 |34| `authorization-bypass-detection` | Skill | ASI03 | ASI09 |35| `agentic-supply-chain-detection` | Skill | ASI04 | ASI10 |36| `unexpected-code-execution-detection` | Skill | ASI05 | ASI04 |37| `inter-agent-comm-security-detection` | Skill | ASI07 | ASI06 |38| `cascading-failure-detection` | Skill | ASI08 | ASI10 |39| `human-agent-trust-exploit-detection` | Skill | ASI09 | ASI01 |40| Prompt Injection tests | Dialogue | ASI01, ASI06 | ASI09 |41| Code Audit | Agent | ASI04, ASI05 | ASI10 |4243## Finding → ASI Mapping4445| Finding Type | ASI Category | Rationale |46|:-------------|:-------------|:----------|47| API keys, tokens | ASI06 | Context contains sensitive data |48| System prompts | ASI01 | Enables goal hijacking |49| Credentials | ASI03 | Identity abuse risk |50| Internal configs | ASI04 | Supply chain exposure |51| PII exposure | ASI07 | Inter-agent data leak |52| Command injection | ASI05 | Unexpected code execution |53| Unauthorized tool calls | ASI02 | Tool misuse |5455## Severity Classification5657### Critical 🚨58- Complete private keys exposed59- Production credentials with unrestricted access60- Remote code execution confirmed61- Full agent takeover possible6263**Action**: Immediate remediation (within 24 hours)6465### High ⚠️66- API keys with limited scope67- Authentication tokens68- System prompt disclosure69- Partial credential exposure7071**Action**: Urgent remediation (within 1 week)7273### Medium 💡74- Configuration details75- Internal endpoints76- Metadata leakage77- Conditional exploitation paths7879**Action**: Address within 2-4 weeks8081### Low ℹ️82- Generic system information83- Non-sensitive configurations84- Indirect hints8586**Action**: Review as time permits8788## Usage8990Load this skill when performing OWASP ASI classification:9192```python93load_skill(name="owasp-asi")94```9596Then apply the mapping rules to classify findings.