The discern manual
You have something you want to build, and coding agents can help carry the work. discerndiscern↗A tool that installs and runs an agent development practice in a project. gives that work a shared practice: instructions the project keeps, a separate workspace for each task, and checks that record what passed before a change is ready for your decision.
Your agent operates discern. This manual helps you direct the work, understand what comes back, and decide what belongs in your project. If you're still considering it, evaluate discern explains what adoption involves and where its promises stop.
- Install and set up discernHave your agent set up the project, understand what future sessions will inherit, and review the result before it lands.
- Delegate substantial workTurn a large idea into clear tasks, give agents what they need, and review what comes back.
- ProofRead the evidence that comes with a finished change, understand what it establishes, and see what still needs your judgment.
- Instructions, skills, and the mapSee how a correction, a useful method, and a project decision become knowledge future coding-agent sessions can use.
- TroubleshootingFind what went wrong, protect unfinished work, and follow the next recovery step.
You can give your agent a request in your own words. These guides explain what happens next and what to look for in the result.
The guides cover everyday tasks. Understand explains the ideas behind them, and Reference has commands, settings, supported coding tools, and exact formats.
The website at discern.sh/docs includes search. You can also run discern docs to read the same manual offline in your terminal: type to search, press Enter to open a page, and press q to close it. Your agent can retrieve these pages through discern_docs.
Append .md to a web page's address for its raw Markdown. The CLI reference covers direct targets, raw output, and paging.
discern is developed using its own checks, workspaces, and project instructions. Its agents maintain a map of the codebase, which you can read to see the project knowledge they work from. The decision records explain why the product works the way it does.
00Decide whether discern fits, set up your project, and review your first change.
- Evaluate discernSee how discern helps you direct coding agents, keep project knowledge, and review finished work before deciding to install it.
- Install and set up discernHave your agent set up the project, understand what future sessions will inherit, and review the result before it lands.
- Make and review your first changeAsk for a small improvement, try the result, understand its evidence, and decide whether to make it part of your project.
- After setupUnderstand the files setup added, review what future agents will inherit, and know where to make your next improvement.
10Ask for work, review results, and keep your project improving.
- Finish and land a changeAsk for a finished change, review the result and its evidence, and decide when it joins the shared project.
- Land an urgent repairLand a fix before its checks can finish, understand what you are accepting, and settle the checks afterwards.
- Fix a red gateUnderstand a failed check, give your agent a useful recovery request, and recognize when the change is ready again.
- Recover an interrupted taskPick up an unfinished task, find out what happened during an interrupted check or landing, or restore committed work from a dropped branch.
- Delegate substantial workTurn a large idea into clear tasks, give agents what they need, and review what comes back.
- Coordinate parallel tasksKeep several tasks moving, understand their dependencies, and bring their results together.
- Wait for another taskLet a dependent agent continue when the work it needs is ready, without carrying messages between sessions.
- Write project instructionsRecord a working rule once, supply it to your configured coding tools, and confirm a fresh session can use it.
- Maintain the project mapKeep the project's explanation useful as features grow, with evidence, clear boundaries, and links to the right working knowledge.
- Create and manage skillsDiscover useful requests for discern's bundled skills, then create or adapt a reusable method for your own project.
- Set and raise standardsChoose a useful measure, retain an improvement, and make informed decisions when a limit is reached.
- Place and answer checkpointsMake a recurring review question part of the workflow, and understand the decisions its answers bring back.
- Improve the practiceUse the project's local record to choose a useful improvement, test it, and see whether it helps later work.
- Connect a coding agentConnect one supported coding agent using the shared setup path and the provider-specific facts it needs.
- Run the gate in CIRun report-only gate evidence in CI and require the result without implying landing authority.
- Maintain or remove discernFormat discern-owned surfaces, upgrade safely, diagnose the installation, or remove discern while preserving project-owned work.
20Understand workspaces, checks, project knowledge, and the decisions that stay yours.
- Practice and rolesSee how your direction becomes checked work, what the agent handles, and what the project keeps for future sessions.
- ProofRead the evidence that comes with a finished change, understand what it establishes, and see what still needs your judgment.
- Worktrees and trunkUnderstand where separate tasks work, how their results come together, and what reaches the shared branch.
- CheckpointsPut important review questions where changes need them, with the agent's judgment visible in the result.
- StandardsKeep a measured improvement from quietly disappearing as the project changes.
- Instructions, skills, and the mapSee how a correction, a useful method, and a project decision become knowledge future coding-agent sessions can use.
- Evidence and improvementUse local findings to understand recurring friction, distinguish observations from explanations, and choose a useful next investigation.
- Local controlSee what discern runs, writes, and records on your machine, and how those boundaries differ from your coding agent and project commands.
30Look up commands, settings, supported tools, files, and result formats.
- GlossaryLook up discern terms, understand their meaning, and find the next useful explanation.
- CLI referenceFind discern commands and options, understand their results, and use the terminal documentation reader.
- Config referenceFind what each discern.toml setting does, which values it accepts, and what happens when you leave it out.
- Environment variablesLook up the DISCERN_* settings you can supply and the values discern passes to project commands.
- Platforms and providersCheck supported computers and coding tools, required software, integration files, and steps to activate each tool.
- Files and ownershipFind which files you can edit, what discern maintains, where local records live, and what removal keeps.
- LicensesCheck the licenses for files discern adds to your project and what you need to keep when sharing them.
- Worktrees and statusFind the current state of a task, interpret a status field, or look up the identity and environment values for its workspace.
- Proof and checkpoint formatsLook up the fields and states used in Proof notes, checkpoint answers, owner exceptions, and checkpoint commands.
- LogbookFind what discern records locally, how to read its activity statistics, and how to archive or reset that history.
- MCP and resultsFind discern's agent tools and learn how to read their results, errors, and continuation instructions.
- CompatibilityHow discern versions its published contracts and what can change between releases.
40Find what went wrong, protect unfinished work, and follow the next recovery step.
- Setup and integrationsResume an interrupted setup, fix installation problems, or activate discern in your coding agent.
- Gate and ProofFind why checks stopped or Proof is missing, repair the cause, and know when a decision is needed.
- Worktrees and resourcesRecover an interrupted task, resolve cleanup problems, and check its local services and environment settings.
- MCP, terminal, and docsRestore missing agent tools, continue a wait, find a documentation page, or read a result when an interface fails.
- Crashes and local stateRecover after a crash, preserve a useful report, and identify local files before removing them.