Webmcp Adapter Creator

v0.1.1

Create fallback site adapters for websites that do not expose native WebMCP. Use when a site needs a new adapter module, tool schema design, browser-side req...

0· 136·1 current·1 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
The skill is for scaffolding and designing browser-side fallback SiteAdapters and includes helper scripts and extensive guidance. The requested files and scripts align with that purpose. One minor mismatch: the SKILL.md prerequisites mention pnpm and Playwright (npx playwright install) and the scripts require tools like perl and ripgrep (rg), but the registry metadata lists no required binaries — the metadata omits some real tooling prerequisites.
Instruction Scope
The SKILL.md explicitly instructs the agent to operate in a logged-in browser session, capture page-originated network requests, prefer in-page execution, and avoid secret replay or credential vaulting. It does not instruct reading unrelated system files or exfiltrating secrets. The workflow is narrowly scoped to adapter design, request discovery, templating, testing, and scaffolding.
Install Mechanism
There is no install spec and this is instruction-plus-scripts only (no network downloads). The included shell scripts are simple file-copy and text-substitution scaffolding and local validation; nothing is downloaded from external URLs and no archives are extracted.
Credentials
The skill does not request environment variables, credentials, or config paths. The documentation explicitly warns against credential vaulting and secret replay. Runtime behavior relies on the browser session (expected) rather than external secrets; this is proportionate to the described adapter tasks.
Persistence & Privilege
always is false and the skill does not request persistent or elevated privileges. It does create scaffold output in the repository tree when run, which is expected for a scaffolding tool and is scoped to its purpose.
Assessment
This skill appears to be what it claims: a scaffolding and guidance pack for building browser-side fallback adapters. Before installing or running scripts, note: (1) the package metadata omits some real prerequisites — you should have pnpm and Playwright browsers (run `npx playwright install`), and the scripts require common CLI tools (perl, ripgrep `rg`, cp, mkdir) to be available; (2) the scaffold script will write files into the repository (packages/adapter-<name> by default) — review the scaffolded output before committing; (3) the SKILL.md explicitly forbids secret replay and vaulting, and the code does not show any network exfiltration, but you should still review the included shell scripts if you will run them in a sensitive environment; (4) if you need the agent to run this autonomously, be aware it can open and control a browser session as part of adapter discovery (expected for this use). If you want higher assurance, verify the existence of the referenced examples/adapter-template and run the validation script in a safe sandbox first.

Like a lobster shell, security has layers — review code before you run it.

latestvk9714tv7fzmpfeg27vzsnsqmwn83gxc5

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Comments