Agent Reach
Security checks across static analysis, malware telemetry, and agentic risk
Overview
This web-research skill is useful but needs review because it asks the agent to use login cookies, run external helpers, bypass anti-bot protections, and potentially publish to social accounts.
Install only if you trust the upstream Agent Reach tooling and are comfortable with external web services. Use separate low-privilege accounts or browser profiles, never provide broad personal browser cookies, inspect remote install steps first, and require explicit approval before any posting, commenting, upload, or anti-bot automation.
Static analysis
No static analysis findings were reported for this release.
VirusTotal
64/64 vendors flagged this skill as clean.
Risk analysis
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.
If the agent or underlying tools use broad browser cookies, they may access services as you and expose or misuse logged-in sessions.
The skill instructs use of browser/session cookies and logged-in platform access, which can act as the user's account identity, without clearly bounding which cookies are used, where they are stored, or how they are protected.
Use `--cookies-from-browser chrome` or configure proxy. ... `Requires login. Use Cookie-Editor to import cookies.` ... `User only provides cookies. Everything else is your job.`
Use a separate browser profile or limited test account, provide only the minimum cookies needed for the specific platform, and review how ~/.agent-reach/ stores or deletes session data.
The agent could post or interact using your account on supported platforms, which can affect your reputation, privacy, or account standing.
The skill goes beyond reading/searching and includes public account-mutating actions, including publishing content and images, without explicit confirmation or reversibility guidance.
Use when: ... `(4) user asks to post, comment, or interact on supported platforms` ... `mcporter call 'xiaohongshu.publish_content(title: "标题", content: "正文", images: ["/path/img.jpg"], tags: ["tag"])'`
Require explicit human confirmation before any post, comment, upload, or account interaction, and verify the exact account, content, images, and destination.
Using anti-bot bypass tooling may violate platform rules, trigger account or IP restrictions, or run helper code whose behavior is not visible in this artifact set.
The instructions explicitly recommend anti-bot bypass automation and running a local helper for WeChat article reading, which is a materially sensitive use of automation even if it supports the stated reading goal.
`Read` (Camoufox — bypasses WeChat anti-bot): `cd ~/.agent-reach/tools/wechat-article-for-ai && python3 main.py "https://mp.weixin.qq.com/s/ARTICLE_ID"`
Avoid anti-bot bypass flows unless you understand the legal and account-risk implications, and inspect the helper code before running it.
Installation behavior and dependencies may change outside this reviewed skill file.
The skill depends on external setup instructions and globally installed tools that are not bundled or pinned in the provided artifact set.
If a channel needs setup (cookies, Docker, etc.), fetch the install guide: `https://raw.githubusercontent.com/Panniantong/agent-reach/main/docs/install.md` ... `npm install -g undici`
Review the upstream install guide and dependency versions before installing or running global package commands.
Queries, URLs, profile lookups, and possibly uploaded content may be visible to third-party services.
The skill sends user-provided URLs, searches, and platform requests to external reader/search/gateway services, which is expected for its purpose but important for privacy.
`curl -s "https://r.jina.ai/URL"` ... `mcporter call 'exa.web_search_exa(query: "query", numResults: 5)'` ... `mcporter call 'linkedin.get_person_profile(...)'`
Do not use the skill with private URLs, confidential topics, or sensitive files unless you trust the external providers involved.