Who it is forA developer with a long, multi-part task who wants the agent to follow a fixed sequence instead of drifting off topic.
The problemLong agent sessions lose focus and skip steps, so the work ends up incomplete or off target.
The outcomeThe task moves through the graded stages in order, with the plan, check-ins, and final review recorded in the session.
What you need first
- DSH CLI installed, version >=0.1.0-rc.6 (the README lists this as the DSH target; DSH is in developer preview with possible breaking changes)
- PowerShell for the README install commands (install.ps1 and the Copy-Item steps)
- git, to clone the repository for the manual install path
- Node.js, to run graded/scripts/measure.mjs (inferred from the node command in the README)
Steps
- Install the packaged plugin with
dsh plugin --profile web add graded/dsh-graded-mode-3.2.0.tgz, since the README names this file and says it can be added directly. (inferred, check the README) - Restart DSH and open a new session.
- Start the brainstorm stage, pick the multiple-choice options to align the goal, then confirm the north star and the written plan.
- Work through each check-in until the group review and final review are complete.
- Read graded/docs/ARCHITECTURE.md if you need the stage rules.
Prompt for your agent
Using the repository yjh051108/dsh-routing-suite, install the graded plugin from graded/dsh-graded-mode-3.2.0.tgz and run the task <task-description> through the graded stages in order. Report the stage reached and the final review result. Verify that each stage produced its output before the next one began.
Caveat. The graded component is shipped as a release candidate (v0.0.1-rc1 for the repo, with a 3.2.0 package file), so behavior may change between versions.
Get 5 new use cases every morning. Free, one short email, unsubscribe anytime.
Get them freeShare: X LinkedIn Hacker News Email Copy link
Disclaimer
Deduped lists open-source repositories that belong to other people. We do not own, maintain, audit, or vouch for any of them, and a repo appearing here is not an endorsement or a safety check. The use cases, steps, and prompts are written by AI and may be incomplete, outdated, or just wrong. Where a use case is marked "Seen in the wild", we link to public posts where someone described something similar, but we have not verified those posts and the use case itself is still our AI's own write-up. Use all of this at your own risk. Only run code, commands, or agents on systems and accounts you are authorized to use, and read the code and every command before you run it, especially anything that touches credentials, files, or the network. Nothing here is professional, security, legal, or financial advice. All trademarks and project names belong to their respective owners, and their use here does not imply affiliation. Content is provided as is, without warranties of any kind. Found a mistake, a safety problem, or something you want removed? Email [email protected] and we will look at it promptly. Full disclaimer page