accessibility-check
v1.0.0审查或改进前端 UI 的语义结构、键盘支持、焦点管理、标签以及常见的 WCAG 相关问题，并将报告保存为 Markdown 文件。当用户提到无障碍、accessibility、a11y、WCAG，或在实现交互组件时自动激活。
⭐ 0· 114·0 current·0 all-time
byBovin Phang@bovinphang
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name, description and SKILL.md consistently describe an accessibility (WCAG 2.1 AA) review and producing a Markdown report. No unrelated credentials, binaries, or installs are requested.
Instruction Scope
The instructions focus on accessibility checks and mandate saving a Markdown report under project-root reports/. They do not attempt to access unrelated system config or external endpoints. The SKILL.md does not explicitly describe how the agent should obtain or analyze project source files (linters, file reads, or tools), so the agent may read workspace files to produce the report — that is expected for a review task but the skill does not document the exact file-read behavior.
Install Mechanism
No install spec and no code files — instruction-only skill. Lowest-risk install profile.
Credentials
No environment variables, credentials, or config paths are requested. The only capability implied is write access to the project workspace to create reports/, which is proportional to the stated purpose.
Persistence & Privilege
always is false and model invocation is allowed (platform default). The skill does not request permanent system presence or modifications to other skills or global agent settings.
Scan Findings in Context
[no_code_files_for_scan] expected: The regex-based scanner found nothing because this is an instruction-only skill (SKILL.md only). That is expected for this kind of skill.
Assessment
This skill appears coherent and low-risk: it will generate an accessibility report and save it under reports/ in the project. Before installing or running it, confirm you are comfortable allowing the agent to read project files (source/markup/CSS) so it can perform the review, and verify where the agent's "Write" tool will create files (it will create reports/accessibility-review-YYYY-MM-DD-HHmmss.md). If you are analyzing sensitive code, consider running the skill on a copy of the repo or in a restricted workspace. Note: the SKILL.md mentions a review tool name ('frontend-craft') as a header only — there are no installs or external network calls specified in the skill.Like a lobster shell, security has layers — review code before you run it.
latestvk972ysm2mj67s0pcxfwfq1j7qn83nttd
License
MIT-0
Free to use, modify, and redistribute. No attribution required.