Security audit
Taku Compact
Security checks for vulnerabilities and agentic risk
Overview
This skill is a disclosed context-briefing utility that reads project/session evidence and writes scoped handoff notes without hidden execution or credential behavior.
Before installing, understand that generated briefs may include excerpts or summaries of repository state, git diffs, commands, user decisions, and session-visible context. Review briefs before sharing them outside the project, especially if the session involved secrets, customer data, or private plans.
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
- 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