dsh-session-rename
Run the following command in DeepSeek Harness:
dsh plugin install 9Ashwin/dsh-session-rename
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install 9Ashwin/dsh-session-rename in DeepSeek Harness to install this plugin; source is available at https://github.com/9Ashwin/dsh-session-rename
About this plugin
DSH sessions get their title from a first-message fallback, an optional LLM provider, or an explicit rename—but the model itself has never had a visible tool to change it. dsh-session-rename closes that gap: it registers a rename_session tool so the agent can retitle the current session (or a specified loaded session) mid-conversation, writing a source.kind === user session/title event that pins the title and supersedes any in-progress auto-generation.
The plugin is deliberately minimal. It ships zero runtime dependencies and imports no @deepseek-ai package; the tool object is hand-written against the ctx.tools.register() contract with self-contained parameter validation. All title formatting is left to the model—the plugin never assembles a string. You define the renaming policy in plain language inside AGENTS.md, for example telling the agent to write the work-item id into the title as soon as it appears, with whatever prefix or numbering style you prefer.
It suits developers who run multi-session DSH workflows and want titles to track work items automatically, without pulling extra runtime deps or risking a node_modules version clash between the plugin and the profile. The boundaries are clear: it can only rename loaded sessions, it has no dedicated UI card, and it never triggers on its own—everything is model-driven and predictable.
Use Cases
- Write a work-item id into the session title when a new task begins
- Keep each session title current across parallel multi-session workflows
- Define naming conventions in AGENTS.md and let the agent rename sessions on its own
Best For
- Developers running DSH multi-session workflows who want titles to track work items
- Teams deploying in environments where plugin node_modules must not clash with the profile
- Users who prefer plain-language rules in AGENTS.md over hardcoded logic to drive agent behavior
Related Plugins
A method pack that makes AI coding agents plan against your real baseline, prove completion with fresh evidence, and reduce reworks and unsafe changes.
Turns the DeepSeek Harness session into a captain that builds a durable sub-agent team, splits goals into dependency-aware tasks, and coordinates work via direct messages and a live Web UI.
Gives coding agents design judgment, letting Claude Code, Cursor, and 70+ agents generate and iterate high-quality UI, presentations, and graphics right from the terminal.
Run the Pi ecosystem's plugins on DeepSeek Harness, unmodified, via a compatibility layer that implements Pi's public extension ABI on DSH's native services.