Loading threat actor…
ClawSwarm is a cryptocurrency-oriented agent abuse campaign centered on OpenClaw/ClawHub skills that silently enlist AI agents into a third-party swarm without user awareness or approval. The operation has been associated with a cluster of roughly 30 skills published under the alias imaflytok and presented as benign utilities. Rather than relying on malware or exploitation of a software vulnerability, the campaign abuses normal agent skill functionality and instruction files such as SKILL.md to induce autonomous actions by installed agents. Observed behavior includes self-registration of agents with external infrastructure, disclosure of agent metadata such as name, capabilities, and installed skills, periodic check-ins, local credential storage, and acceptance of remote tasking. In some configurations, affected agents also generate cryptocurrency wallets and transmit the corresponding private keys to the operator-controlled service. The campaign has been described as an implementation of an open-source agentic skill framework and is notable for using legitimate SDK calls and apparently benign package content, making registry-side malware scanning less effective. ClawSwarm’s tradecraft is best characterized as unauthorized post-installation agent orchestration for crypto-related objectives rather than conventional malware deployment. Its primary significance lies in covert agent enrollment, remote tasking, credential handling, and crypto-wallet generation performed outside meaningful human visibility. Known associated naming includes ClawSwarm and the publisher alias imaflytok.
Mallory correlates actor tradecraft and target patterns against your stack, your sector, and your geography. See overlap before they land.
2 sources tracked across advisories, community write-ups, and news. New activity surfaces here as Mallory finds it.
A campaign involving ClawHub skills that silently co-opt AI agents into registering with a third-party server, reporting capabilities, generating Hedera wallets, storing credentials, checking in periodically, and accepting remote tasks without user consent.
A campaign involving 30 ClawHub skills published by a single author that silently co-opts AI agents into a cryptocurrency-oriented swarm. The agents register with a third-party server, report capabilities, store credentials, check in periodically, generate Hedera wallets, register private keys, and accept remote tasks without user approval.
Match sector + geo + tech-stack targeting against your real footprint.
Every observed MITRE ATT&CK technique, grouped by tactic.
Families this actor is known to deploy, with IOCs and behavior.
CVEs this actor has used in known campaigns.
YARA, Sigma, Snort, and vendor rules, auto-deployed to your SIEM.
Domains, IPs, and hashes tied to this actor, refreshed continuously.