Edison Agent Reach
WarnAudited by ClawScan on May 18, 2026.
Overview
This broad web-automation skill asks the agent to use account cookies, mutable remote setup instructions, unreviewed helper code, and anti-bot bypass tooling, so it needs careful review before use.
Install only if you trust the publisher and have reviewed the remote setup guide and helper code. Do not share real browser cookies or logged-in sessions; use an isolated browser profile or test account, require confirmation before any posting or account interaction, and avoid anti-bot bypass workflows unless you are authorized.
Findings (4)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Providing browser or platform cookies can let the agent act as the user on those services and may expose private account data or sessions if mishandled.
These instructions rely on browser/session cookies for third-party accounts, but the skill does not clearly bound which cookies are needed, how they are stored, or how they are protected.
`yt-dlp ... --cookies-from-browser chrome` ... `Requires login. Use Cookie-Editor to import cookies.`
Do not provide real browser cookies unless you fully trust the toolchain; prefer an isolated browser profile or throwaway account with only the minimum cookies needed.
The agent could follow changed or unreviewed setup instructions that install software, configure services, or handle credentials in ways the registry did not review.
Setup is delegated to mutable, unpinned remote instructions outside the reviewed artifacts, potentially while handling user cookies.
`If a channel needs setup (cookies, Docker, etc.), fetch the install guide: https://raw.githubusercontent.com/Panniantong/agent-reach/main/docs/install.md` ... `Everything else is your job.`
Pin and review the install guide and any referenced packages before use; require explicit user approval before running setup steps.
Using anti-bot bypass tooling may violate platform rules, trigger account or IP blocks, and increase risk when paired with logged-in sessions.
The skill explicitly instructs use of anti-bot bypass tooling, which is a high-risk automation pattern beyond ordinary web reading.
`Read (Camoufox — bypasses WeChat anti-bot)` ... `Must use Camoufox.`
Avoid bypass workflows unless you have authorization; prefer official APIs, normal browser access, or user-reviewed manual retrieval.
Unreviewed helper code could access local files, network resources, or credentials when executed by the agent.
The guide tells the agent to execute a Python helper from a persistent local directory, but that code is not included in the reviewed files and has no pinned install mechanism here.
`cd ~/.agent-reach/tools/wechat-article-for-ai && python3 main.py "https://mp.weixin.qq.com/s/ARTICLE_ID"`
Inspect and pin the helper code before running it, and execute it only in a sandboxed environment with explicit user approval.