Security audit
Security checks for vulnerabilities and agentic risk
This skill is a local creative-idea generator with no evidence of hidden network access, credential use, persistence, or destructive behavior.
Install this if you want a mostly Chinese, local brainstorming helper. Expect rule-based suggestions and rough scores, not validated business or market analysis; review outputs before using them professionally.
The stated purpose is a local creative ideation assistant, but the file also includes operational behaviors such as compiling Python files, running tests, and checking package consistency. This description-behavior mismatch is dangerous because it obscures the true execution surface of the skill, making reviewers and users less likely to anticipate code execution or local environment interaction.
The skill metadata does not declare any tool restrictions, yet the repository description and validation steps indicate file-reading and shell execution capabilities via Python compilation and test commands. In an agent environment, undeclared capabilities weaken least-privilege guarantees and can allow the skill to perform actions beyond what a user would reasonably expect from a creativity tool.
The trigger phrase "突破创作瓶颈" is a generic expression a user might naturally say when asking for help, without necessarily intending to invoke this specific skill. The file lists trigger phrases but does not provide exclusion conditions or narrower activation context to distinguish routine conversation from deliberate skill activation.
This code returns user-facing titles, descriptions, labels, and suggestions entirely in Chinese across multiple response paths. The file does not provide any opt-in, locale selection, or documented justification for enforcing a single language, which conflicts with the policy against forcing a specific language without user choice.
subprocess module calls execute external commands. Without careful input validation, this enables command injection.
raise SystemExit("SKILL.md frontmatter version mismatch")
print("[verify] compiling")
subprocess.run([sys.executable, "-m", "py_compile", "handler.py", "scripts/test.py"], cwd=ROOT, check=True)
print("[verify] testing branches")
subprocess.run([sys.executable, "scripts/test.py"], cwd=ROOT, check=True)
subprocess module calls execute external commands. Without careful input validation, this enables command injection.
subprocess.run([sys.executable, "-m", "py_compile", "handler.py", "scripts/test.py"], cwd=ROOT, check=True)
print("[verify] testing branches")
subprocess.run([sys.executable, "scripts/test.py"], cwd=ROOT, check=True)
print("[verify] ok")
The manifest description uses Chinese text exclusively, which can impose a specific language on users without any opt-in or indication that the skill is intended only for a Chinese-speaking or region-specific audience. The stated policy requires flagging language or locale constraints when they are forced without user choice or clear justification.
The package description is written entirely in Chinese, which signals a language-specific experience but provides no indication that users can choose another language or that the skill is intentionally limited to a Chinese-speaking audience. This can violate language/locale policy when a skill appears to impose a specific language without opt-in.
The test cases include multiple hard-coded Chinese strings such as themes, keywords, and ideas, which indicates the skill behavior is being validated primarily in a specific language. Because there is no surrounding comment or logic offering language choice or documenting a justified locale restriction, this can conflict with the policy against forcing a specific language without user opt-in.