Back to skill

Security audit

meta-deep-research

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed research workflow with a local learning-state helper, with no evidence of exfiltration or destructive intent.

Install only if you are comfortable with a Chinese-language research helper that may create research files and keep local learning history. Run the learner only against trusted skill directories, avoid elevated privileges, and keep research outputs in a dedicated workspace.

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)

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/learner.py:61
Finding

Symlink-Following Arbitrary File Overwrite in Learning-State Storage

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (7)

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The skill claims a sophisticated research, verification, orchestration, and self-evolving workflow, but the finding indicates the actual implemented behavior is materially different and centers on local persistence of learning statistics and preferences. This mismatch is dangerous because users and orchestrators may grant trust, autonomy, or access based on the declared purpose while the real behavior performs undisclosed stateful data retention, undermining transparency and consent.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
84% confidence
Finding

The skill advertises behavior that writes research artifacts such as outline.yaml, fields.yaml, results/item_N.yaml, and report.md, but it declares no explicit tool scope or permissions boundaries. In agent environments, undeclared file-write capability increases the risk of unintended or overly broad filesystem modifications, making the skill harder to review and constrain safely.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The manifest description is written entirely in Chinese and the rest of the skill documentation also presents the skill in Chinese, with no indication that users may choose another language. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation unless the locale restriction is clearly documented and justified.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

This code file contains natural-language descriptions and CLI usage text entirely in Chinese, and later emits Chinese-only user-facing messages. Under the language/locale policy, forcing a specific language without user opt-in is a policy concern unless the locale restriction is explicitly justified.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The script prints status, insight, and reflection messages exclusively in Chinese during normal execution. This enforces a single locale on users without presenting a choice or documenting a justified regional limitation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
92% confidence
Finding

This markdown file contains user-facing natural language exclusively in Chinese and does not mention that the language is optional, user-selected, or required for a region-specific purpose. Under the policy, forcing a specific language without opt-in can be a locale policy violation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
82% confidence
Finding

The JSON stores the capability name as the Chinese phrase "蒸馏增强" with no accompanying indication that the skill supports multiple languages or that it is intentionally region-specific. Under the policy for natural-language violations, a fixed locale-specific label can be a concern when no opt-in or justification is provided.

Content

No source excerpt is available for this finding.