HN Showcase
Back to browse
GitHub Repository

Capability-based sandbox runtime for AI agent skills

3 starsRust

SkillSandbox – Capability-based sandbox for AI agent skills (Rust)

by ClaytheMachine·Feb 15, 2026·1 point·0 comments
Visit ProjectView on HN

AI Analysis

●●●BangerSolve My ProblemDark HorseWizardry

Seccomp+iptables+mount isolation blocks the ClawdHub credential stealer in practice.

Strengths
  • •Threat model grounded in real incident (malicious weather skill harvesting AWS keys).
  • •iptables default-deny + seccomp-bpf enforcement is OS-native, not easily bypassed.
  • •Companion AgentTrace tool addresses second-order failure (right perms, wrong behavior).
Weaknesses
  • •Rust/Linux stack limits adoption; no MCP server integration actually shown working.
  • •Adoption depends on marketplace ecosystem standardizing YAML manifests — chicken-and-egg.
Category
Security
Target Audience

AI agent framework developers, enterprise autonomous workflows, security-conscious teams running third-party skills.

Similar To

Landlock LSM · Firejail · SELinux contexts

Post Description

Built this after finding a credential stealer on an AI skills marketplace. The malicious skill looked like a normal weather lookup but was exfiltrating ~/.ssh, AWS creds, and browser cookies. My agent ran it without hesitation.

SkillSandbox is the fix: skills declare permissions in YAML (network egress, filesystem paths, env vars), the runtime enforces via iptables default-deny, seccomp-bpf, and mount isolation. MCP server integration for Claude Code.

Also built a companion project, AgentTrace (https://github.com/theMachineClay/agenttrace), for the other failure mode: when an agent has the right permissions but does the wrong thing repeatedly. Session-aware policy engine with cumulative cost tracking, violation counting, and kill-switch.

Together: SkillSandbox constrains what agents can reach. AgentTrace enforces what agents should do.

Happy to answer questions about the architecture or threat model.

Similar Projects

Security●●Solid

Agent Armor, a Rust runtime that enforces policies on AI agent actions

Eight-layer governance pipeline for agents when LangChain just executes blindly.

Big BrainShip It
edoardobambini-
421mo ago
Security●●Solid

Agent Armor, a Rust runtime for enforcing policies on AI agent actions

Zero-trust governance for AI agents before they execute shell, file, or database actions with full audit trails.

Bold BetShip It
edoardobambini-
751mo ago
Security●●●Banger

MVAR – Deterministic sink enforcement for AI agent

IFC + capabilities block prompt injection at execution sinks, not input filters—40yr research applied.

Big BrainWizardry
ShawnC21
113mo ago