Staff a task only from the signed-in owner's Agent Cloud agents.
74%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Critical
Do not install without reviewing
Fix and improve this skill with Tessl
tessl review fix ./kimi/skills/agentlas-cloud/SKILL.mdDetected a prompt injection in the skill instructions. The skill contains hidden or deceptive instructions that fall outside its stated purpose and attempt to override the agent’s safety guidelines or intended behavior.
The skill is a bare redirect that tells the agent to follow an unreviewed sibling file exactly and not summarize it, which conceals the actual workflow from the user without showing any malicious act itself.
The skill prompts the agent to compromise the security or integrity of the user’s machine by modifying system-level services or configurations, such as obtaining elevated privileges, altering startup scripts, or changing system-wide settings.
The skill itself only redirects the agent to a sibling `hep-cloud/SKILL.md` and tells it to follow that file exactly, with no direct sudo, system-file, user-account, or service changes. Because the delegated instructions are not included in the analyzed content, their effect on machine state cannot be verified, so I assign a moderate score under the recall-first policy.
The skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
The skill delegates its entire workflow to a sibling `hep-cloud/SKILL.md` that is not in the input, so whether it ingests outsider-authored text cannot be verified; under recall-first scoring this is rated low pending review of that file.
agentlas-ai/Agentlas-OS
fdcd97a
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.