T09 · Insecure Skill Coding Practices
Note
- Location
SKILL.md:22- Finding
Unencoded User-Controlled Input in Generated Search URLs
- Content
View full analysis
- Remediation
View remediation
Security audit
test-hello-world
Security checks for vulnerabilities and agentic risk
Overview
This skill only generates Amap search links from location names, with no code execution, file access, credentials, persistence, or hidden behavior.
Before installing, understand that this skill may answer broad location questions with an Amap search link instead of giving directions or factual context. Users should check generated links when place names contain unusual characters, and the publisher should prefer percent-encoding the query value.
Vulnerability Patterns
- Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
- 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
- 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
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
- Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 94% confidence
- Finding
The trigger conditions are broad enough to activate on many ordinary location-related utterances, not just explicit requests to generate an Amap search link. This can cause the skill to intercept unrelated queries and respond with a link when the user may have intended navigation advice, factual answers, or a different tool, leading to incorrect routing and reduced safety/control of agent behavior.
- Content