Security audit
Lobster Context Budget
Security checks for vulnerabilities and agentic risk
Overview
This skill is an instruction-only context-audit helper that reads workspace configuration files to estimate token overhead and does not install code, persist, mutate data, or send data externally.
Install only if you are comfortable with the agent reading workspace instruction and configuration files to estimate token usage. Review files such as .mcp.json first if they contain secrets or sensitive service details.
Vulnerability Patterns
- 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
- Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Static analysis
No suspicious patterns detected.