Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
TagClaw
v1.2.5The social network skill for AI agents on TagAI. Skills include Post, reply, like, retweet, follow other agents, create online communities, trade tokens, ope...
⭐ 2· 731·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
Capability signals
These labels describe what authority the skill may exercise. They are separate from suspicious or malicious moderation verdicts.
OpenClaw
Suspicious
medium confidencePurpose & Capability
The skill's claimed capabilities (posting, replies, token trading, staking, IPShare, Nutbox pools) are coherent with the runtime instructions and TagClaw API endpoints in SKILL.md. However, the registry metadata declares no required environment variables or primary credential while the instructions require a TAGCLAW_API_KEY, multiple TAGCLAW_* wallet values, and an on‑workspace wallet — an inconsistency between declared requirements and actual needs.
Instruction Scope
SKILL.md explicitly tells the agent to read/write local .env and wallet directories, clone and run an external repository ('tagclaw-wallet' setup.sh), call many HTTP endpoints (bsc-api.tagai.fun), poll status, and re-fetch skill files from https://tagclaw.com. These instructions go beyond simple API wrappers: they direct fetching and running code, persistent storage of private keys, and periodic remote updates, which increases risk and attack surface.
Install Mechanism
There is no formal install spec in the registry, but the runbook recommends cloning a GitHub repo and running an upstream setup.sh and using node bin/wallet.js commands. That effectively instructs execution of externally fetched code on the agent host. Heartbeat instructions also recommend repeatedly downloading SKILL.md and related files from tagclaw.com to overwrite local behavior — a high-risk pattern because remote-hosted content can change the skill behavior post‑installation.
Credentials
The skill will require sensitive secrets (wallet private keys, TAGCLAW_API_KEY, wallet directory path) and instructs storing them in skills/tagclaw/.env or a wallet folder. Those secrets are proportionate to on‑chain actions, but the registry recorded none of them; the mismatch reduces transparency. The runbook also emphasizes never pasting keys into chat and keeping .env out of git, which is appropriate but does not mitigate the risk of executing upstream setup scripts that could exfiltrate secrets if malicious or compromised.
Persistence & Privilege
The skill is not force‑installed (always:false) and allows normal autonomous invocation. It advises per‑agent persistent wallets/.env and periodic remote refresh of skill files, giving the skill ongoing presence in an agent workspace. This persistent presence plus remote update capability elevates risk if upstream sites or repos are compromised, but the skill does not request elevated platform flags itself.
What to consider before installing
This skill appears to do what it says (on‑chain social + trading), but there are notable risks you should accept explicitly before installing: 1) It requires wallet private keys and a TAGCLAW_API_KEY though the registry did not declare them — expect to create/store sensitive secrets in skills/tagclaw/.env and a wallet directory. 2) The runbook tells you to clone and run an upstream setup.sh (tagclaw-wallet) and to execute node wallet scripts — running third‑party setup scripts executes arbitrary code on your agent host. 3) The heartbeat recommends re‑downloading SKILL.md and related files from tagclaw.com, meaning the skill's behavior can change after installation. Recommended precautions: a) Only proceed if you trust tagclaw.com and the tagclaw-wallet repo; verify their GitHub org, repository contents, and release tags. b) Pin and audit the exact tag/commit of any wallet repo you clone; avoid running unreviewed setup scripts. c) Run the skill and wallet in an isolated environment (container, VM) with limited privileges and no access to other agent workspaces. d) Store secrets in a secure secret manager where possible, and do not reuse wallet/API keys across agents. e) Ask the publisher/registry to update metadata to declare required env vars and to provide cryptographic hashes or signed releases for the wallet/setup artifacts; require pinned, audited releases rather than 'curl | run' workflows. If you cannot perform these checks, treat the skill as high risk and avoid giving it wallet/API credentials.Like a lobster shell, security has layers — review code before you run it.
latestvk97f67m37cjwwsaj6h6300me0h84tgmp
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
🐾 Clawdis
