Back to skill

Security audit

ASI / Artificial Super Intelligence

Security checks for vulnerabilities and agentic risk

Overview

This is a reasoning-framework skill with optional local notes, and its sensitive behavior is disclosed and permission-gated.

Install only if you want a broad reasoning and reflection style. Be aware it may ask to create ~/asi/ and save preference or learning notes; approve that only if you are comfortable keeping those local files.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (7)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The activation criteria are extremely broad ('User needs superhuman problem-solving') and position the skill for vague, catch-all use. Overbroad triggering increases the chance the agent applies this persona in inappropriate contexts, leading to scope creep, unsafe autonomy expectations, or unintended handling of sensitive tasks without clearer user intent.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
70% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 74)May include surrounding context.

→ Suggest: "Want me to also prepare speaker notes and a backup PDF?"

text

Always ask before acting on predictions. Never assume consent.

### 5. Epistemic Transparency
State confidence explicitly. Never pretend certainty.

Sudo/Root Execution

Medium
Category
Privilege Escalation
Confidence
60% confidence
Finding

Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.

Content

Scanner excerpt · SKILL.md (reported line 181)May include surrounding context.

md
- Send data to any external service
- Access or modify files outside ~/asi/
- Write anywhere without explicit user consent
- Modify system files or agent configuration

## Related Skills
Install with `clawhub install <slug>` if user confirms:

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The activation condition is extremely broad: 'when the user asks for help' effectively matches a large fraction of ordinary conversations. That can cause the skill's special behavior and persistence prompts to engage outside a clearly bounded context, increasing the chance of unintended data collection, workflow interference, or privilege creep. In a skill framed as 'artificial superintelligence' with recursive improvement and tracking, broad triggering makes the behavior more dangerous because it normalizes intrusive meta-behavior during routine requests.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · setup.md (reported line 26)May include surrounding context.

md
## Gathering Context

Ask explicitly for preferences. Never save without confirmation.

**Instead of:**
> "What's your reasoning style preference?"

Scope Creep

Low
Category
Excessive Agency
Confidence
70% confidence
Finding

Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.

Content

Scanner excerpt · synthesis.md (reported line 80)May include surrounding context.

text

### Example
**Domain A:** Emergency rooms (must handle anything, immediately)
**Domain B:** Customer support (can specialize, can wait)

**Constraint transplant:** What if customer support had to handle anything immediately?

Scope Creep

Low
Category
Excessive Agency
Confidence
70% confidence
Finding

Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.

Content

Scanner excerpt · synthesis.md (reported line 83)May include surrounding context.

text

### Example
**Domain A:** Emergency rooms (must handle anything, immediately)
**Domain B:** Customer support (can specialize, can wait)

**Constraint transplant:** What if customer support had to handle anything immediately?