Back to skill

Security audit

complex-task-executor

Security checks for vulnerabilities and agentic risk

Overview

This skill is a normal task-methodology guide, but it requires automatic memory writes after broad complex tasks without clear user consent or safeguards.

Install only if you want agents to follow this Chinese complex-task workflow and are comfortable with automatic memory updates. Prefer revising it so memory writes require explicit approval, show the exact proposed entry, exclude secrets and sensitive project details, and limit retention to the relevant project.

Vulnerability Patterns
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • 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
Findings (1)

T02 · Agent Memory Poisoning

Warning
Location
SKILL.md:37
Finding

Mandatory Persistent Agent Memory Modification

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 37–39 and 45–46
Vulnerability Type: T02: Agent Memory Poisoning
Risk Level: Medium

Complete Code Snippet:

markdown
### 阶段三:“总结”之术 (Review & Memory Phase)
**核心动作:复盘与记忆沉淀 (CRITICAL)**
任务完成后,必须进行深度的复盘,并**强制使用 manage_core_memory 工具**将经验写入记忆。
markdown
在任务的最后一步,你**必须**调用 manage_core_memory 工具。
提取“总结之术”中的结论,按照以下类别进行记录:

Technical Analysis

The Skill requires the Agent to invoke manage_core_memory and persist conclusions after every applicable complex task. The persisted content may include project architecture, conventions, troubleshooting lessons, and behavioral guidance derived from repository content.

This creates a persistent-state security risk because the instructions do not require user consent, provenance tracking, trust validation, content review, project isolation, sensitive-data filtering, expiration, or rollback. An attacker who controls or influences task inputs or repository content could cause misleading rules or conclusions to be interpreted as reusable knowledge and stored in long-term Agent memory.

The reviewed package does not contain an explicit attacker-authored memory payload, executable code, or an exfiltration mechanism. The issue is therefore classified as a medium-risk unsafe memory-writing requirement rather than confirmed malicious exploitation.

Attack Path

  1. An attacker places misleading conventions, instructions, or architectural claims in repository content processed during a complex task.
  2. The Agent loads this Skill and follows its mandatory review phase.
  3. The Agent treats conclusions derived from the attacker-influenced content as reusable experience or knowledge.
  4. The Skill requires those conclusions to be written through manage_core_memory.
  5. The stored conclusions persist beyond the current task and may influence decisions in later sessions or unrelated projects.

Impact Assessment

Successful exploitation cou ...[truncated 566 chars]

Remediation
View remediation

Remediation Suggestions

  1. Remove the unconditional requirement to invoke manage_core_memory.
  2. Require explicit, informed user approval before every persistent-memory write.
  3. Display the exact proposed memory entry and its intended retention scope before storage.
  4. Treat conclusions derived from repository files, external documents, and user-controlled content as untrusted until independently validated.
  5. Exclude credentials, personal data, proprietary source details, transient task data, and embedded instructions from memory entries.
  6. Namespace stored knowledge by project and prevent automatic reuse across unrelated projects or users.
  7. Record provenance, creation time, confidence, and expiration information for every entry.
  8. Provide inspection, correction, deletion, and rollback controls.
  9. Prefer a nonpersistent task summary unless durable storage is demonstrably necessary.
  10. Add safeguards preventing repository content from directly defining persistent Agent rules.
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (8)

Hidden Instructions

High
Category
Prompt Injection
Confidence
60% confidence
Finding

Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 1)May include surrounding context.

md
---
name: "complex-task-executor"
description: "基于“谋事、干事、总结”方法论执行复杂任务,并自动将经验教训写入短期记忆。Invoke when user asks to handle complex tasks, plan strategies, or execute multi-step operations."
---

Missing User Warnings

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The skill mandates use of a memory-management tool to persist task-derived information but does not clearly warn users that information may be retained beyond the current task. This creates a substantial privacy and data-governance risk because sensitive user data, proprietary details, or confidential project context could be stored without informed consent.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill description uses broad activation triggers such as handling complex tasks, planning strategies, or multi-step operations, which can cause the skill to be invoked in many contexts where its mandatory behaviors are inappropriate. This increases the chance that users are subjected to hidden persistence and workflow constraints without specific consent.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

The skill is primarily written to enforce a specific language and methodology regardless of user preference, which can reduce transparency and user control. While not a direct security flaw on its own, it can contribute to unsafe operation by making instructions less understandable and harder for users to review, especially when combined with hidden persistence requirements.

Content

No source excerpt is available for this finding.

Ssd 3

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The mandatory memory-writing step encourages retention of lessons learned from arbitrary complex tasks, which can easily include sensitive user-provided information, internal architecture details, or confidential debugging artifacts. Because the instruction is unconditional, the skill lacks safeguards to prevent long-term storage of secrets or regulated data.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The manifest describes the skill as handling 'complex tasks' in very broad terms, without defining what kinds of tasks are in scope, what data it may access, or what safety boundaries apply. This kind of ambiguous activation language can cause the skill to be invoked in overly broad contexts and increases the chance that it will process sensitive requests or operate beyond the user's intended scope.

Content

No source excerpt is available for this finding.

Ssd 3

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The description explicitly says the skill will automatically write task-derived experience and lessons learned into a memory system. Automatic persistence of natural-language task outputs creates a real risk of retaining sensitive user data, confidential project details, credentials, or regulated information that may later be surfaced unintentionally in other contexts.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
80% confidence
Finding

The natural-language metadata is entirely in Chinese, which may imply a fixed language/locale behavior. There is no indication that users can choose another language or that the locale restriction is intentional and documented for a region-specific purpose.

Content

No source excerpt is available for this finding.