Agent Reach Skill · Web access
Route the web request to the backend that can actually serve it.
Panniantong’s MIT Agent Reach package provides a multi-platform routing Skill and CLI for supported web backends. It documents safe/default, dry-run, and explicit system-install modes plus backend health checks; login, API keys, installed tools, and platform coverage still vary by the route you choose.
Agent Reach is the MIT-licensed Skill and CLI in the Panniantong/Agent-Reach repository. Panniantong is the repository owner, Neo Reid is named as the package author, and Agent Eyes appears in the LICENSE copyright notice. It routes read requests across documented platform backends, exposes health and setup checks, and distinguishes default, dry-run, and explicit system-install paths.
Built on proven foundations
The original expertise stays intact. Focused refinements make it easier to start, follow the guidance, and know when the work is finished.
Original foundation
The original Skill knowledge, package structure, attribution and original license terms are preserved. Follow the source link to see where this edition began.
View the original projectAntigravity optimization
A source-faithful edition with the improvements explained below. The download includes the original attribution and everything needed to follow the Skill’s guidance.
More useful from the first prompt
Keep the depth you came for. Get clearer starting points, practical guidance and an explicit finish line.
Start a web research request with the right route while keeping read-only and dedicated-skill boundaries in view.
Set up platform/backend discovery, reference routing, non-empty verification, and limitation reporting before using a backend.
Carry platform, backend, command, and login/version limits alongside a result for the next decision or reviewer.
Choose among documented web and platform backends for the request instead of hard-coding one access path.
Inspect whether the selected route, local tool, or login state is ready before asking it to fetch data.
Keep the routing instructions connected to the package and CLI that provide the actual platform adapters.
Distinguish safe/default, dry-run, and explicit system-install choices before changing the local setup.
Route a read request through the backend that has the needed platform support and credentials.
Use the documented health or doctor path to find missing tools, login state, or configuration.
Keep a platform’s backend and limitations visible while an agent gathers information.
# Agent Reach This page is bound to the MIT-licensed Panniantong/Agent-Reach repository at commit a19a171fa980a0785849596492e0af4db800c82f. Panniantong owns the repository, Neo Reid is named in pyproject.toml as package author, and the captured LICENSE carries Agent Eyes copyright. The source includes a routing Skill, CLI/package configuration, and platform references. Its documented backends cover multiple platforms, but access depends on the selected route, local tools, login state, API keys, and platform availability. The useful path is identify the platform → inspect the backend or doctor state → install/configure the required route → perform the read request → report the boundary. Safe/default, dry-run, and explicit system-install modes remain distinct. This fixture records the source identity and runtime disclosure without claiming that a backend is live here.
Use these phrases to activate this skill in your AI coding assistant:
