tabilet-skills
Run the following command in DeepSeek Harness:
dsh plugin install tabilet/tabilet-skills
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install tabilet/tabilet-skills in DeepSeek Harness to install the plugin, with the source repository at https://github.com/tabilet/tabilet-skills.
About this plugin
Switching between DeepSeek Harness, Codex, and Claude Code often means tracking the same project milestones, task states, and acceptance records in three different formats. tabilet-skills treats one canonical Markdown source as the single source of truth, so all three tools read the same project ledger without format fragmentation.
The plugin ships seven workflow skills (Propose, Reconcile, Goal, Upgrade, Init, Archive, and more), each gated behind a preview and explicit approval before any write. The Memory Bank sidebar aggregates six read-only views—Overview, Tasks, Acceptance, Memory, History, and Compatibility—with a five-second refresh cycle, no embedded content execution, and no model requests, keeping project records safe and auditable at all times.
Best suited for developers or small teams who collaborate across multiple AI coding tools and need a consistent engineering ledger. Repositories already organised in the tabilet project format can plug in directly and use all skills and panels without any migration step.
Use Cases
- View project milestones, tasks, and acceptance status in DSH without switching tools
- Preview workflow requests via Propose or Reconcile, then execute only after explicit approval
- Memory Bank panel renders engineering memory and history read-only with a five-second refresh and zero model requests
Best For
- AI coding developers working across DSH, Codex, and Claude Code
- Small teams that need a unified Markdown engineering ledger
- Engineering leads who prefer approval gates over auto-submitted workflows
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.