Back to skill

Security audit

Agent Spec

Security checks for vulnerabilities and agentic risk

Overview

This skill is a documentation template for designing bounded AI agents and does not itself run tools, access data, or install persistence.

This skill is reasonable to install if you want help drafting agent specifications. Review any separately linked companion skills before using them, and ensure the agent specs it produces include explicit approval gates, budgets, and escalation rules for real-world actions.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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)

Unbounded Resource Access

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill allows unbounded resource consumption (API calls, storage, compute). Without rate limits or quotas, a compromised or misbehaving agent can cause denial-of-service or cost overruns.

Content

Scanner excerpt · SKILL.md (reported line 40)May include surrounding context.

md
| search_kb | read context | yes | none |
| send_email | notify | **no** | **human approval** |

**3. Control loop** — plan → act → observe → reflect; the stopping condition; and a hard **max-steps / max-cost budget** so it can't loop forever.

**4. Guardrails & approval gates** — which actions require a human yes (default: anything irreversible, outbound, or spending), input/output validation, and allow/deny lists. Pair irreversible actions with a dry-run preview (see [`action-runner`](../action-runner/SKILL.md)).