dsh-evidence-task-board
Run the following command in DeepSeek Harness:
dsh plugin install MkaliezZ/dsh-evidence-task-board
Paste the following prompt into your AI chat to install this plugin:
Install this plugin in DeepSeek Harness by running dsh plugin install MkaliezZ/dsh-evidence-task-board; the source repository is available at https://github.com/MkaliezZ/dsh-evidence-task-board .
About this plugin
In multi-step agent workflows on DeepSeek Harness, task state tends to live only in each agent's ephemeral context. Once a session is interrupted, the context window overflows, or work must resume across sessions, progress becomes hard to audit or pick up where it left off. dsh-evidence-task-board addresses this gap by providing a set of persistent, deterministic task-state primitives for DSH, so every task's lifecycle is explicitly recorded rather than implicitly held in memory.\n\nCore capabilities include: three-phase task management covering creation, status transitions, and evidence attachments; a stable content-addressed SHA-256 board digest that lets you verify board consistency across renders; and a human-readable board rendering designed for consoles and logs. Persistence is explicitly supplied by the host application. The plugin does not invent hidden autonomous tasks and does not execute any external tools—it is responsible only for state logic and presentation.\n\nIt is well suited for teams building multi-step workflows on DSH that need cross-session auditability and replay. If you want every state transition to be traceable, every digest to be verifiable, and every render to be readable—without introducing additional autonomous behavior—this lightweight plugin is a practical fit.
Use Cases
- Reconstructing and auditing task progress after a workflow interruption
- Generating a verifiable SHA-256 digest for the current task board state
- Presenting task creation, status transitions, and evidence in a human-readable format
Best For
- Developers building multi-step agent workflows on DeepSeek Harness
- Teams that need cross-session auditability and replay capability
- Architects who prefer explicitly persisted task state over in-memory retention
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.