dsh-watcher
aa2246740
Read-only Agent work-path observer for DeepSeek Harness
mokuyoaxis254Updated 3d ago
Make destructive AI-agent actions reversible by default — quarantine + audit + human escalation for rm/git destructive operations. Reliability infrastructure, not a sandbox.
Practical score
84 / 100
Setup
Ready to use
License
MIT
Language
Python
Copy the command, or hand the whole task to your harness agent.
git clone https://github.com/mokuyoaxis/agent-guard.gitPlease install the DeepSeek Harness plugin "agent-guard" for me. - GitHub repository: mokuyoaxis/agent-guard (https://github.com/mokuyoaxis/agent-guard) - Type: a Cordis plugin (installed into a DSH profile) - It should work out of the box with no extra config…
A cleaned excerpt from the project's own documentation.
AGENT-GUARD Disclosure checks · Recoverable deletion · Agent experiments · 简体中文 Agent Guard has three main capabilities: Exfil-guard checks text before sending it, Delete-guard preserves recovery data before supported destructive actions, and Guard-lab tests agent and host behaviour through controlled experiments. A shared Python Core supports the command-line tools and optional host adapters.…
Open on GitHubSimilar tools from the same corner of the ecosystem.
aa2246740
Read-only Agent work-path observer for DeepSeek Harness
InfinitePersistence
Unofficial DeepSeek Harness plugin for board serial console, logging, and model-visible interaction.
alibaba-damo-academy
Turn your AI agent app into an interactive ideation and design workbench. A general-purpose, open-source framework with editable panels and versioned projects across DSH, Codex, and other agent hosts.