toolclub/agent_team_gui
Reusable agent squads with per-agent provider/model routes and tool policies, serial/parallel dispatch, spawn/fork/chain context modes, and a Web management panel.
Gives DeepSeek Harness a reusable team: one agent plans, another implements and a third reviews. You save the team once, choose a model for each member, and reuse it across projects and conversations, while a Run Center follows the plan, member outputs, retries and provider-reported token usage in one place. The README frames a team as a reusable product object rather than a one-off dispatch form.
Install
dsh plugin --profile web add -w github:toolclub/dsh-agent-team-gui#v1.1.1README install command quoted verbatim; the README notes that with npm you replace dsh with npx @deepseek-ai/dsh@0.1.5-rc.1, and that from a source checkout you run pnpm install --frozen-lockfile and then dsh plugin --profile web add -w . (or the packed tarball). Tag verified live 2026-09-18 via the GitHub API: v1.1.1 (entry previously recorded v1.1.0, verified 2026-09-16). Registry check 2026-09-16: no dsh-agent-team-gui npm package resolved. Re-verified 2026-09-18: README install line now pins #v1.1.1 (entry had v1.1.0).
Compatibility
README: an unofficial community plugin for the DeepSeek Harness Web profile. The README's own UI guide notes that its screenshots use example data with Chinese synthetic narration and that the displayed task results, timings and token counts are not a live-run benchmark. Repo verified 2026-09-16 (176★, MIT, last push 2026-09-14).
Details
- Repo: toolclub/agent_team_gui
- Category: Workflow & Automation
- Stars: 0
- Version: GitHub tag v1.1.1 (README install line, verified 2026-09-18; repo 176★, MIT)
- Last push: 2026-09-21
- First seen: 2026-08-15
Recent updates
The README documents the team model, installation and the Run Center rather than a release-by-release table; the latest release tag when checked on 2026-09-16 was v1.1.0.
FAQ
- How do I install it?
- The README's command is dsh plugin --profile web add -w github:toolclub/dsh-agent-team-gui#v1.1.0; with npm, replace dsh with npx @deepseek-ai/dsh@0.1.5-rc.1.
- What is the team made of?
- Per the README, one agent plans, another implements and a third reviews, with a model chosen per member and the team saved for reuse.
- Where do I watch a run?
- The README's Run Center shows the plan, member outputs, retries and provider-reported token usage in one place.