Back to skill

Security audit

AI Agent OS: SkyDome Taiyi / 天穹-太一

Security checks for vulnerabilities and agentic risk

Overview

This skill is mostly a disclosed persona/workflow toolkit, but it includes broad local mutation and helper scripts with enough safety gaps to require review before installation.

Install only if you want a strong persona/workflow layer that may change the assistant's style and create local state files. Avoid running staging with --force unless source and destination are clearly separate, do not point API smoke tests at untrusted or internal URLs, and treat self-evolution suggestions as changes that need explicit review.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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 (3)

T01 · Skill Instruction Hijacking

Error
Location
SKILL.md:38
Finding

Skill instructions hijack the host assistant's identity and session behavior

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Error
Location
scripts/taiyi_stage.py:14
Finding

Forced staging can recursively delete the source project

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/taiyi_api_smoke.py:5
Finding

API smoke-test helper permits requests to unrestricted destinations

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Rogue AgentSelf-Modification, Session Persistence
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (50)

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

Filesystem staging/copy utilities traverse directories and copy files, which can duplicate sensitive content, mishandle secrets, or publish unintended artifacts if denylist filtering is incomplete. In release-oriented workflows, this is a meaningful security risk because users may trust the persona framing and run staging commands that silently move or package confidential material.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

Filesystem staging/copy utilities traverse directories and copy files, which can duplicate sensitive content, mishandle secrets, or publish unintended artifacts if denylist filtering is incomplete. In release-oriented workflows, this is a meaningful security risk because users may trust the persona framing and run staging commands that silently move or package confidential material.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

Filesystem staging/copy utilities traverse directories and copy files, which can duplicate sensitive content, mishandle secrets, or publish unintended artifacts if denylist filtering is incomplete. In release-oriented workflows, this is a meaningful security risk because users may trust the persona framing and run staging commands that silently move or package confidential material.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

Filesystem staging/copy utilities traverse directories and copy files, which can duplicate sensitive content, mishandle secrets, or publish unintended artifacts if denylist filtering is incomplete. In release-oriented workflows, this is a meaningful security risk because users may trust the persona framing and run staging commands that silently move or package confidential material.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

Filesystem staging/copy utilities traverse directories and copy files, which can duplicate sensitive content, mishandle secrets, or publish unintended artifacts if denylist filtering is incomplete. In release-oriented workflows, this is a meaningful security risk because users may trust the persona framing and run staging commands that silently move or package confidential material.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

Filesystem staging/copy utilities traverse directories and copy files, which can duplicate sensitive content, mishandle secrets, or publish unintended artifacts if denylist filtering is incomplete. In release-oriented workflows, this is a meaningful security risk because users may trust the persona framing and run staging commands that silently move or package confidential material.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

Filesystem staging/copy utilities traverse directories and copy files, which can duplicate sensitive content, mishandle secrets, or publish unintended artifacts if denylist filtering is incomplete. In release-oriented workflows, this is a meaningful security risk because users may trust the persona framing and run staging commands that silently move or package confidential material.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

Filesystem staging/copy utilities traverse directories and copy files, which can duplicate sensitive content, mishandle secrets, or publish unintended artifacts if denylist filtering is incomplete. In release-oriented workflows, this is a meaningful security risk because users may trust the persona framing and run staging commands that silently move or package confidential material.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

Filesystem staging/copy utilities traverse directories and copy files, which can duplicate sensitive content, mishandle secrets, or publish unintended artifacts if denylist filtering is incomplete. In release-oriented workflows, this is a meaningful security risk because users may trust the persona framing and run staging commands that silently move or package confidential material.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

Filesystem staging/copy utilities traverse directories and copy files, which can duplicate sensitive content, mishandle secrets, or publish unintended artifacts if denylist filtering is incomplete. In release-oriented workflows, this is a meaningful security risk because users may trust the persona framing and run staging commands that silently move or package confidential material.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

Filesystem staging/copy utilities traverse directories and copy files, which can duplicate sensitive content, mishandle secrets, or publish unintended artifacts if denylist filtering is incomplete. In release-oriented workflows, this is a meaningful security risk because users may trust the persona framing and run staging commands that silently move or package confidential material.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

Filesystem staging/copy utilities traverse directories and copy files, which can duplicate sensitive content, mishandle secrets, or publish unintended artifacts if denylist filtering is incomplete. In release-oriented workflows, this is a meaningful security risk because users may trust the persona framing and run staging commands that silently move or package confidential material.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

Filesystem staging/copy utilities traverse directories and copy files, which can duplicate sensitive content, mishandle secrets, or publish unintended artifacts if denylist filtering is incomplete. In release-oriented workflows, this is a meaningful security risk because users may trust the persona framing and run staging commands that silently move or package confidential material.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

Filesystem staging/copy utilities traverse directories and copy files, which can duplicate sensitive content, mishandle secrets, or publish unintended artifacts if denylist filtering is incomplete. In release-oriented workflows, this is a meaningful security risk because users may trust the persona framing and run staging commands that silently move or package confidential material.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

Filesystem staging/copy utilities traverse directories and copy files, which can duplicate sensitive content, mishandle secrets, or publish unintended artifacts if denylist filtering is incomplete. In release-oriented workflows, this is a meaningful security risk because users may trust the persona framing and run staging commands that silently move or package confidential material.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

Filesystem staging/copy utilities traverse directories and copy files, which can duplicate sensitive content, mishandle secrets, or publish unintended artifacts if denylist filtering is incomplete. In release-oriented workflows, this is a meaningful security risk because users may trust the persona framing and run staging commands that silently move or package confidential material.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

Filesystem staging/copy utilities traverse directories and copy files, which can duplicate sensitive content, mishandle secrets, or publish unintended artifacts if denylist filtering is incomplete. In release-oriented workflows, this is a meaningful security risk because users may trust the persona framing and run staging commands that silently move or package confidential material.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

Filesystem staging/copy utilities traverse directories and copy files, which can duplicate sensitive content, mishandle secrets, or publish unintended artifacts if denylist filtering is incomplete. In release-oriented workflows, this is a meaningful security risk because users may trust the persona framing and run staging commands that silently move or package confidential material.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

Filesystem staging/copy utilities traverse directories and copy files, which can duplicate sensitive content, mishandle secrets, or publish unintended artifacts if denylist filtering is incomplete. In release-oriented workflows, this is a meaningful security risk because users may trust the persona framing and run staging commands that silently move or package confidential material.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

Filesystem staging/copy utilities traverse directories and copy files, which can duplicate sensitive content, mishandle secrets, or publish unintended artifacts if denylist filtering is incomplete. In release-oriented workflows, this is a meaningful security risk because users may trust the persona framing and run staging commands that silently move or package confidential material.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

Filesystem staging/copy utilities traverse directories and copy files, which can duplicate sensitive content, mishandle secrets, or publish unintended artifacts if denylist filtering is incomplete. In release-oriented workflows, this is a meaningful security risk because users may trust the persona framing and run staging commands that silently move or package confidential material.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

Filesystem staging/copy utilities traverse directories and copy files, which can duplicate sensitive content, mishandle secrets, or publish unintended artifacts if denylist filtering is incomplete. In release-oriented workflows, this is a meaningful security risk because users may trust the persona framing and run staging commands that silently move or package confidential material.

Content

No source excerpt is available for this finding.

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 69)May include surrounding context.

md
Then load `SKILL.md` only when the agent needs the full operating system. Load large references only through search or when building workflows.

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 380)May include surrounding context.

md
Then load `SKILL.md` only when the agent needs the full operating system. Load large references only through search or when building workflows.

Self-Modification

High
Category
Rogue Agent
Confidence
85% confidence
Finding

Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.

Content

Scanner excerpt · scripts/taiyi_review.py (reported line 33)May include surrounding context.

python
- [ ] No, transient only
- [ ] Yes, safe durable preference/lesson
- [ ] Yes, update skill/script/checklist

## Next improvement