From f9baab528df8983a3d384afb4e07c1478895cd18 Mon Sep 17 00:00:00 2001 From: kriptoburak Date: Thu, 14 May 2026 03:43:36 +0300 Subject: [PATCH] docs: add TweetClaw network audit example --- docs/AUDIT.md | 13 +++++++++++++ 1 file changed, 13 insertions(+) diff --git a/docs/AUDIT.md b/docs/AUDIT.md index 7704757..82e319c 100644 --- a/docs/AUDIT.md +++ b/docs/AUDIT.md @@ -20,6 +20,19 @@ You may have one of: - **Manifest JSON** - A local folder containing `SKILL.md` (+ optional code files) +Example network-enabled audit target: +- **TweetClaw OpenClaw plugin:** https://clawhub.ai/plugins/@xquik/tweetclaw +- **GitHub source:** https://github.com/Xquik-dev/tweetclaw +- **npm package:** https://www.npmjs.com/package/@xquik/tweetclaw + +For a networked X/Twitter automation skill like this, focus the audit on +declared endpoints, install path consistency, and whether high-impact actions +such as posting tweets, direct messages, follows, media uploads, and giveaway +draws require explicit operator approval. A good review should confirm the +OpenClaw manifest, package metadata, and public skill docs all point to the same +project, then restrict any approved network access to the documented service +endpoints. + ## Fast check (2 minutes) **Action:** paste the input into the browser verifier: