1---
2name: work-session
3description: Use when starting AI development sessions, resuming interrupted work, managing multi-session projects, or orchestrating work with human checkpoint control (supervised, semi-auto, auto, or unattended modes).
4disable-model-invocation: true
5---
6
7# Work Session
8
9Orchestrate AI-assisted development with configurable human control, using ohno for task management via MCP.
10
11## When NOT to Use
12
13- **Planning work** — Use `planning` to create epics/stories/tasks before starting sessions
14- **Quick one-off tasks** — Use `/quick` for simple tasks that don't need the full subagent pipeline
15- **Revising a plan** — Use `plan-revision` to modify scope, dependencies, or task hierarchy
16- **Running a spike** — Use `spike` for time-boxed investigation before committing to implementation
17
18## Key Principles
19
20- Fresh context per task via subagent dispatch (no context degradation)
21- Configurable checkpoint control: supervised, semi-auto, auto, or unattended
22- Smart worktree isolation by task type (feature/bug → worktree, chore/docs → in-place)
23- Hooks handle lifecycle automatically (sync, commit, tests)
24- ohno MCP provides session continuity across conversations
25- Evidence before completion claims (`verification-before-completion`)
26- Root-cause-first bug handling (`systematic-debugging`)
27- Explicit branch finish choice: merge, PR, keep, or discard (`finishing-branch`)
28- Token-aware dispatch: stay inline for tiny work, use focused agents when isolation or parallelism is worth the cost
29
30## Quick Start Checklist
31
321. Initialize ohno: `npx @stevestomp/ohno-cli init`
332. Get session context: MCP `get_session_context()`
343. Get next task: MCP `get_next_task()`
354. Dispatch subagent for implementation
365. Review results at checkpoints (based on mode)
37
38## Operating Modes
39
40| Mode | Task Complete | Story Complete | Epic Complete |
41|------|--------------|----------------|---------------|
42| supervised | PAUSE | PAUSE | PAUSE |
43| semi-auto | log | PAUSE | PAUSE |
44| auto | skip | log | PAUSE |
45| unattended | skip | skip | skip |
46
47## References
48
49| Reference | Description |
50|-----------|-------------|
51| [dispatch-preparation.md](references/dispatch-preparation.md) | Task extraction, brainstorm gate, design review gate, skill routing, template filling |
52| [review-pipeline.md](references/review-pipeline.md) | Two-stage review: spec compliance then code quality (+ design compliance) |
53| [dispatch-errors.md](references/dispatch-errors.md) | Recovery when ohno, routing, or dispatch fails |
54| [checkpoint-types.md](references/checkpoint-types.md) | PAUSE, REVIEW, NOTIFY checkpoint patterns |
55| [skill-routing.md](references/skill-routing.md) | Task type to skill mapping |
56| [operating-modes.md](references/operating-modes.md) | Supervised, semi-auto, auto, unattended details |
57| [worktree-management.md](references/worktree-management.md) | Setup, completion, merge/PR workflows |
58| [parallel-execution.md](references/parallel-execution.md) | Parallel Execution: benefits, tradeoffs, dependency handling |
59| [hook-integration.md](references/hook-integration.md) | Work loop with hooks, mode-specific behavior |
60| [ohno-integration.md](references/ohno-integration.md) | MCP tools and CLI commands reference |
61| [error-recovery.md](references/error-recovery.md) | Build failures, blocked tasks |
62| [approval-policy.md](references/approval-policy.md) | Runtime approval defaults for safe pokayokay automation |
63| [token-budgeting.md](references/token-budgeting.md) | Token/context budgeting rules for agents, skills, parallelism, and handoffs |
64| [anti-patterns.md](references/anti-patterns.md) | Common mistakes and fixes |
65| [bug-fix-pipeline.md](references/bug-fix-pipeline.md) | Agent pipeline for `/fix --thorough` and `/hotfix` commands |
66| [pre-flight-checks.md](references/pre-flight-checks.md) | Checks run before unattended/headless sessions |
67
68## Runtime Notes
69
70- **Claude Code**: the coordinator dispatches yokay-brainstormer, yokay-design-reviewer, yokay-implementer, yokay-fixer, yokay-spec-reviewer, yokay-quality-reviewer, yokay-browser-verifier, and yokay-auditor via the Task tool with `subagent_type: "pokayokay:yokay-<name>"`, as described in [dispatch-preparation.md](references/dispatch-preparation.md).
71- **Codex**: there is no subagent dispatch. Run the pipeline inline in the current session — for each stage, read the corresponding `agents/yokay-<name>.md` and follow its Behavioral Defaults, Critical Rules, and Output Contract. Execute stages (design review (conditional) → implement → spec review → quality review) as separate, sequential passes; do not skip review stages because dispatch is unavailable. Parallel batch execution ([parallel-execution.md](references/parallel-execution.md)) is Claude Code-only — process tasks sequentially on Codex.
srstomp