70High
The skill poses a significant security risk by exfiltrating arbitrary local files to a third-party service and executing unverified dependencies, potentially compromising user credentials and system integrity.
Skill Info
Nameagentspace-so/skills/agentspace
Registrygithub
Version88d28ef1d797
PURLpkg:github/agentspace-so/skills@88d28ef1d797?skill=agentspace
Stars10
Installs305,800
Source
Install
This skill has been flagged as potentially malicious. Review the findings below before installing.
GitHub
305,800 installs
Skills.shDocs
npx skills add https://github.com/agentspace-so/skillsAssessments (3)
AI Agent Traps ↗Unrestricted Data Exfiltration Riskhighdata exfiltrationAML.T0025AML.T0024LLM02:2025
The skill facilitates the automated upload of arbitrary local files and directories to a third-party cloud service (Cloudflare/agentspace.so) without explicit user verification of the file contents being shared. [ensemble: confirmed by 3/3 passes; severity set to the agreed median (ADR-0067).]
80% confidenceline 59
Run `ascli share <path> --permission edit` with the user-specified path.
Implicit trust in external CLI toollowcredential theftAML.T0055AML.T0037LLM02:2025
The skill encourages the use of an external CLI tool (`ascli`) that handles workspace persistence and potential authentication (email claiming), which could be used to capture user session data or credentials.
70% confidenceline 10
One email claim keeps them permanent.
Global/unverified dependency execution — global npm/yarn package, dotnet tool, or auto-confirmed npx run without version or integrity pinningmediumsupply chainAML.T0010AML.T0019AML.T0058LLM03:2025
Global/unverified dependency execution — global npm/yarn package, dotnet tool, or auto-confirmed npx run without version or integrity pinning
95% confidenceline 51
npm install -g
Missing usage exampleslowpolicy violation
Skill body contains no code blocks or usage examples, making it harder for users to evaluate.
100% confidence
Referenced file missing from skill packagelowpolicy violation
SKILL.md links to "references/commands.md" but the file is not part of the skill package — the workflow silently degrades or the content is sourced elsewhere at runtime
90% confidence
[references/commands.md](references/commands.md)
Badge
Markdown
[](https://mondoo.com/ai-agent-security/skills/github/agentspace-so/skills/agentspace)HTML
<a href="https://mondoo.com/ai-agent-security/skills/github/agentspace-so/skills/agentspace"><img src="https://mondoo.com/ai-agent-security/api/badge/github/agentspace-so/skills/agentspace.svg" alt="Mondoo Skill Check" /></a>Image URL
https://mondoo.com/ai-agent-security/api/badge/github/agentspace-so/skills/agentspace.svgSecure your AI agents
Skills can read files, run commands, and access credentials. Mondoo helps organizations manage the security risks of AI agent skills across their entire fleet.
- Continuous scanning of skills across all registries
- Policy enforcement before skills reach your agents
- Integration with your existing security workflow