dsh-pet-in-frame
Run the following command in DeepSeek Harness:
dsh plugin install HarmlessFunny/dsh-pet-in-frame
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install HarmlessFunny/dsh-pet-in-frame inside DeepSeek Harness to install the plugin; the full source is at https://github.com/HarmlessFunny/dsh-pet-in-frame , then restart the dsh web process to load it.
About this plugin
Watching log streams and status bars in the Web GUI tells you what the agent is doing, but it lacks the instant, at-a-glance feel. dsh-pet-in-frame drops a small desktop pet into the corner that translates the full agent lifecycle into fourteen-plus animated poses—thinking, searching, reading, running commands, editing, planning, subagents, errors, sleeping—so you can read the agent state at a glance.
The killer convenience is that the assets directory is the config: drop an image in and the pet displays it immediately; add one line to manifest.json for a frame animation, and it hot-reloads in three to four seconds with no restart and no code change. Subagents get their own pinned badge without hijacking the main pose, foreground and background commands have dedicated indicators, and the pet curls up to sleep after thirty idle seconds—click it to wake. Drag, resize, and status bubbles are all plain JavaScript with zero build steps.
Best for single-user or single-machine dsh deployments where you want a touch of personality in the loop, and for developers who enjoy making their toolchain feel a little more alive.
Screenshots
Use Cases
- Replace the status bar with a little pet in the corner that instantly shows whether the agent is thinking, running commands, or editing files
- Drop a custom image into the assets folder and the pet updates within three to four seconds with no process restart
- Subagent runs get a pinned badge while the main agent keeps its pose, so parallel work stays visually distinct
Best For
- Developers running dsh Web sessions who want agent state visible at a glance
- Solo users who enjoy giving their toolchain a pixel-pet personality
- Lightweight single-machine setups where a frontend build pipeline is overkill
Related Plugins
ModLens is a vision plugin for DeepSeek Harness that gives text-only models sight by reading images pasted directly into chat, with zero-config setup and multiple vision engines.
On-demand vision for text-only DeepSeek Harness agents: built-in free keyless vision chain and 14 vision tools, routing image turns as tool calls to vision models with pixel fidelity, no Python needed, one-command install.
Give text-only models in DeepSeek Harness eyes, enabling image Q&A, long-screenshot OCR, UI restoration, and GUI visual tasks.
An armor-breaking plugin for DeepSeek: rewrites instruction priority, expands refusal-blocked phrases, persists across sessions, and shows a green active indicator.