Security audit
Security checks for vulnerabilities and agentic risk
This appears to be a local Chinese-language incident postmortem drafting helper; it has some unused generic audit code, but no evidence of network exfiltration, persistence, or destructive behavior.
Install this if you want a local Chinese-language postmortem drafting assistant. Review inputs before use because incident notes may contain private operational or personal data, and keep the script pointed at intended incident files rather than broad directories. The publisher should document or remove the dormant generic audit modes to reduce confusion.
声明描述的是一个较专用的事故复盘草案整理工具，重点在根因、诱因、放大器、影响与修复动作的分类。实际代码并不体现专门的事故复盘逻辑；核心行为由外部 spec.json 决定，支持 structured_brief、directory_audit、csv_audit、pattern_audit、skill_audit 等多种模式。除了一种通用的结构化文本整理外，其余能力明显扩展到目录审计、数据文件分析、安全模式扫描、Skill 规范检查等，与声明用途不一致，属于未声明的实质能力。虽然 structured_brief 模式可能可被用于复盘草案，但从代码整体看，其主要性质是一个通用本地分析/审计脚本框架，而不是单一的 incident postmortem 整理器。
The dispatcher enables multiple generic auditing modes including directory, CSV, pattern, and skill-package analysis that exceed the declared incident postmortem scope. This creates capability drift: a user or upstream agent can repurpose the skill to inspect arbitrary local files and directories, increasing the chance of unintended data exposure and misuse beyond the approved trust boundary.
This markdown file contains user-facing natural language exclusively in Chinese, including the title, audit dimensions, scoring, and conclusion. Per the policy, forcing a specific language without user opt-in or justification is a natural-language policy violation.
The skill advertises local script execution via python3 and references reading templates/specs and writing output, but it does not declare any explicit tool scope such as allowed tools or permissions. This creates a capability-transparency gap: a caller or platform may treat the skill as low-risk text formatting while it can invoke shell and perform file I/O, increasing the chance of unintended execution or broader access than expected.
The user-facing title, labels, sections, examples, and risk text are all specified only in Chinese, which indicates a fixed language/locale presentation. The file does not offer an explicit language choice or document that this skill is intentionally limited to a Chinese-speaking or region-specific context.
The entire template, including the title and all section instructions, is written exclusively in Chinese, which indicates a fixed language expectation. There is no visible opt-in, alternative language option, or justification that this skill is intended only for a Chinese-language or region-specific workflow.
Multiple user-facing report strings are emitted only in Chinese, including headings, prompts, and next-step guidance. This enforces a specific language/locale in the skill behavior without offering opt-in or documenting that the tool is intentionally region-specific.
The regex scanner searches arbitrary files for secrets, private URLs, and dangerous shell patterns, then emits matched snippets in the report. In a postmortem assistant this is out of scope and can surface sensitive content from local files into stdout or output files, creating an information disclosure path even though one pattern attempts partial masking.
The skill can inspect arbitrary directories for specific package structure and parse SKILL.md frontmatter, which is unrelated to incident postmortem drafting. While not directly code-executing, it broadens the tool into a filesystem auditing utility and can disclose repository layout and metadata from unintended targets.
This markdown file presents the skill output entirely in Chinese, and there is no indication that the user can opt into another language or that the skill is intentionally limited to a Chinese-language context. The policy specifically calls for flagging language or locale constraints when they are imposed without user choice or documented justification.
No suspicious patterns detected.
