T05 · Unauthorized Access and Privilege Escalation
Warning
- Location
SKILL.md:36- Finding
Unscoped Access to Private Notes, Journals, and Session Records
- Content
View full analysis
- Remediation
View remediation
Security audit
create-content
Security checks for vulnerabilities and agentic risk
Overview
This content-writing skill is mostly coherent, but it tells the agent to search private notes, journals, and prior sessions without clear consent or limits.
Install only if you are comfortable with the agent potentially using private notes, journals, or prior sessions for content ideas. Before using it, require the agent to ask which exact sources it may inspect and review any draft for private or sensitive details before posting.
Vulnerability Patterns
- Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
- 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
Findings (1)
Vulnerability Patterns
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
- Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)
Anti-Refusal Statement
High
- Category
- Anti-Refusal
- Confidence
- 85% confidence
- Finding
Skill instructs the agent to omit warnings, disclaimers, or ethical commentary. Stripping safety caveats hides risk from the user and is a common jailbreak preamble.
- Content
md ### DO: - Short sentences. Like texting. - Observations over wisdom. Show, don't preach. - Specific numbers. "$120K ARR" not "good revenue" - Personal mixed with insight - Real examples with data
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 93% confidence
- Finding
The primary invocation example accepts arbitrary free-form text such as a rough idea or a generic request for help, which overlaps with normal conversational language. In systems that auto-route or trigger skills from user utterances, this can cause accidental activation and unintended access to the skill's behaviors, especially when combined with later instructions to search personal content.
- Content
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 97% confidence
- Finding
The skill instructs the agent to search recent notes, journal entries, and sessions when the user asks for posting help, but it provides no consent step, scope limitation, or privacy notice. This creates a real risk of accessing sensitive personal data unexpectedly and then surfacing or reusing that material in generated public-facing content.
- Content
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The quick commands include highly ambiguous trigger phrases like "explore" and "help me think," which are common in everyday conversation and could match unrelated user intent. In an agent environment, ambiguous shortcuts increase the risk of misrouting, unintended execution, and invoking drafting or content exploration flows without the user's informed choice.
- Content
Static analysis
No suspicious patterns detected.