# Persona Workflows

Use these paths when you want to operate Open Orchestra by role instead of
learning the full command catalog first. Commands assume the installed CLI form:
`orchestra`.

## First Visible Value

Goal: complete one local workflow run and preview release readiness before
learning every role-specific artifact.

Commands:

```bash
orchestra init
orchestra health --json
orchestra task add --id DEMO-001 --title "Ship a governed README update" --owner developer --paths "README.md"
orchestra workflow run --task DEMO-001 --gates none
orchestra status
orchestra release candidate --dry-run --json
```

Expected artifacts:

- `.agent-workflow/tasks.json` contains `DEMO-001`.
- `.agent-workflow/workflow-runs.jsonl` contains the completed lifecycle.
- Release candidate dry-run output shows what is ready and what evidence is
  still missing.

Recovery:

- Run `orchestra health --json` if local tools are missing.
- Use `orchestra workflow runs` to find a paused or failed run.
- Move to the production paths below when the story needs human gates,
  explicit sizing, command evidence, QA review, and release readiness approval.

## Product Owner Refinement

Goal: turn a backlog item into a task that is ready for architecture and
delivery.

Commands:

```bash
orchestra health --json
orchestra task add --id STORY-001 --title "..." --owner product_owner \
  --paths "src,docs,test" --goal "..." --acceptance "..." \
  --risks "..." --test-strategy "..."
orchestra github sync --issue 123 --task STORY-001 --comment --dry-run --json
orchestra context --task STORY-001
```

Expected artifacts:

- `.agent-workflow/tasks.json` task entry with backlog item, acceptance
  criteria, assumptions, risks, paths, and test strategy.
- GitHub sync dry-run or comment evidence when using GitHub issues.
- Context output showing memory, missing artifacts, and selected workflow.

Common blockers:

- Missing acceptance criteria or test strategy.
- GitHub CLI unavailable in local shell.
- Issue tracker is not GitHub.

Recovery:

- Add missing readiness fields with `orchestra task update --id STORY-001 ...`.
- Use `orchestra github sync --transport mcp-skill` only as a runtime-adapter
  boundary; tracker MCP skills for GitHub, Jira, Bitbucket, GitLab, or custom
  trackers must inject equivalent issue/comment/close operations.
- Continue locally with task metadata when tracker sync is unavailable.

Related docs: [first-use flow](../README.md#quick-start),
[GitHub sync](orchestra-mvp.md), and [memory hooks](orchestra-mvp.md).

## Developer Execution

Goal: implement a scoped task with role handoffs, evidence, and workflow gates.

Commands:

```bash
orchestra delegation decide --task STORY-001
orchestra workflow run --task STORY-001 --gates phase
orchestra decision add --task STORY-001 --owner architect \
  --title "Story sizing" --decision "m 3 points" \
  --context "..." --consequences "..." --status accepted
orchestra memory hook --point before_implementation --task STORY-001 \
  --role developer --json
orchestra evidence add --task STORY-001 --role developer --type command \
  --summary "npm run precommit passed" --command "npm run precommit" \
  --exit-code 0
```

Expected artifacts:

- Workflow run records in `.agent-workflow/workflow-runs.jsonl`.
- Decisions and handoffs under `.agent-workflow/decisions/` and
  `.agent-workflow/handoffs/`.
- Command evidence under `.agent-workflow/evidence/`.

Common blockers:

- Architect sizing missing before development.
- Active locks on overlapping paths.
- Context budget too high or memory output too broad.

Recovery:

- Record the sizing decision printed by the workflow failure.
- Inspect conflicts with `orchestra graph plan --json`.
- Query role-scoped memory with `orchestra memory query --role developer`.

Related docs: [autonomous workflow](autonomous-workflow.md),
[runtime flow](runtime-llm-flow.md), and [memory governance](orchestra-mvp.md).

## QA Validation

Goal: prove acceptance criteria with traceable evidence and release-ready
coverage.

Commands:

```bash
orchestra playwright plan --task STORY-001
orchestra playwright evidence --task STORY-001 --kind screenshot \
  --summary "Browser AC passed" --path path/to/screenshot.png
orchestra review --task STORY-001 --role qa --result approve \
  --findings "AC covered" --recommendation "Release handoff approved"
orchestra gate --gate qa-release --task STORY-001
```

Expected artifacts:

- QA review in `.agent-workflow/reviews/`.
- Unit, integration, manual, or Playwright evidence mapped to acceptance
  criteria.
- QA gate result showing no missing plan, execution status, or QA evidence.

Common blockers:

- Browser/UI AC has only report evidence and no Playwright artifact.
- QA review blocks are unresolved.
- Evidence summary does not identify the AC it covers.

Recovery:

- Attach screenshot, trace, or video evidence for browser criteria.
- Resolve blocker reviews with a new QA review after fixes.
- Use evidence summaries that name the acceptance criterion or say
  "covers all acceptance criteria."

Related docs: [release readiness](orchestra-mvp.md) and
[Playwright evidence](orchestra-mvp.md).

## Tech Lead Oversight

Goal: keep parallel work coordinated and reviewable.

Commands:

```bash
orchestra graph plan --json
orchestra review --task STORY-001 --role tech_lead --result approve \
  --findings "Diff and tests reviewed" --recommendation "Proceed"
orchestra pr-summary --task STORY-001
orchestra commands manifest --json
```

Expected artifacts:

- Graph plan with ready, blocked, locked, and dependency-ready tasks.
- Tech lead review artifact with severity and recommendation.
- PR summary with evidence, risks, and remaining gaps.

Common blockers:

- Multiple agents edit the same path without locks.
- Command behavior is duplicated outside the manifest.
- PR lacks command evidence or reviewer context.

Recovery:

- Split work by stable path boundaries and record locks.
- Check the manifest before adding commands.
- Add missing evidence and rerun `orchestra pr-summary --task STORY-001`.

Related docs: [command contracts](command-contracts.md),
[traceability](traceability-flow.md), and [runtime adapters](runtime-adapters.md).

## Release Management

Goal: decide whether a story or candidate release can ship.

Commands:

```bash
orchestra release check --json
orchestra release evidence --kind smoke --summary "smoke passed"
orchestra release evidence --kind rollback --summary "rollback verified"
orchestra release evidence --kind post-release --summary "post-release smoke passed"
orchestra release evidence --kind release-notes --summary "release notes published"
orchestra gate --gate release-readiness --task STORY-001
orchestra release candidate --version 0.5.0-beta.0 --json
```

Expected artifacts:

- Release check JSON with acceptance coverage, blockers, locks, smoke, and
  rollback gaps.
- Smoke and rollback evidence events.
- Candidate release plan under `.agent-workflow/releases/`.

Common blockers:

- Uncovered acceptance criteria.
- Active locks or blocker reviews.
- Missing smoke or rollback evidence.
- Dirty working tree during release check.

Recovery:

- Attach required evidence or record an accepted risk on the task.
- Resolve locks and blocking reviews.
- Run `orchestra upgrade --smoke --json` after installing a new package version.

Related docs: [upgrade dogfooding](orchestra-mvp.md),
[package naming](package-naming.md), and [release readiness](orchestra-mvp.md).

## Technical Writer Review

Goal: keep public documentation, site copy, command examples, release notes, and
help content accurate for the intended audience.

Commands:

```bash
orchestra workflow phase-plan --task STORY-001 --json
orchestra doc-sync audit --task STORY-001 --json
orchestra review --task STORY-001 --role technical_writer --result approve \
  --findings "Docs, links, commands, and generated site content reviewed" \
  --recommendation "Documentation ready for release"
```

Expected artifacts:

- Technical Writer review in `.agent-workflow/reviews/`.
- Documentation diff or link evidence for updated docs, README, changelog,
  release notes, runbooks, help content, or public site copy.
- Site QA evidence when generated site content, diagrams, visible copy, or
  responsive layout changes.

Common blockers:

- Public site copy is hardcoded in React instead of generated from
  `docs/site-manifest.json`.
- Command examples were not validated against the CLI, command manifest, or
  build output.
- `docs/architecture.md` and `site/public/architecture.mmd` describe different
  diagrams.

Recovery:

- Update the source markdown or manifest, then run `npm run site:content`.
- Run `npm run site:build` and capture screenshots for visible site changes.
- Use `orchestra doc-sync audit --task <id> --json` to find missing
  documentation coverage.

Related docs: [public site content workflow](site-content-workflow.md),
[adoption guide](adoption-guide.md), and [release matrix](release-test-matrix.md).

## Docs Review Checklist

Before release, verify:

- Persona paths use `orchestra`, not local `node bin/orchestra.js`, unless the
  section is explicitly about repository development.
- Package references use `@jterrats/open-orchestra`.
- Links cover first-use onboarding, web console, memory, GitHub sync, release
  readiness, and upgrade dogfooding.
- Commands list expected artifacts, common blockers, and recovery steps.
- Site copy reflects implemented capabilities, not planned-only features.
