openup-quick-task
Fast iteration mode for small changes - simplified workflow with minimal overhead
Best use case
openup-quick-task is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Fast iteration mode for small changes - simplified workflow with minimal overhead
Teams using openup-quick-task should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/openup-quick-task/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How openup-quick-task Compares
| Feature / Agent | openup-quick-task | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Fast iteration mode for small changes - simplified workflow with minimal overhead
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# Quick Task - Fast Iteration Mode
**Quick Task** is a lightweight workflow for small changes and rapid iteration. It combines multiple steps into a single command while maintaining essential OpenUP practices.
## When to Use
Use Quick Task for:
- Small bug fixes (< 50 lines changed)
- Documentation updates
- Configuration changes
- Quick experiments
- Hot fixes
**Target**: Complete tasks in 50% less time than standard workflow.
## When NOT to Use
Do NOT use for:
- New features (use standard workflow)
- Major refactoring (use `/openup-start-iteration`)
- Tasks requiring architecture review (use full team)
- Multi-hour development work
## Process
### 1. Quick Context Load
```bash
# Load minimal context only
python3 .claude/scripts/batch-context.py --minimal
```
### 2. Quick Branch (optional)
If not skipping branching:
```bash
# Detect trunk and create quick branch
BRANCH_NAME="quick/$(date +%Y%m%d-%H%M%S)-$(echo $task | tr ' ' '-' | head -c 20)"
git checkout -b $BRANCH_NAME
```
Then initialize iteration state on the **quick** track — this is the same `quick` track `/openup-start-iteration` selects for tiny scopes (see [tracks.md](../../../docs-eng-process/tracks.md)). The quick track only requires the `log_written` and `roadmap_synced` gates — there is no plan or team gate:
```bash
python3 scripts/openup-state.py init \
--task-id "{task_id or generated id}" \
--iteration 0 \
--phase construction \
--track quick \
--branch "$(git rev-parse --abbrev-ref HEAD)" \
--worktree "$(git rev-parse --show-toplevel)" \
--force
```
(If branching is skipped and no `.openup/state.json` will exist, skip this and the gate steps below — quick tasks remain lightweight.)
### 3. Execute Task
Implement the change:
- Read task description
- Make necessary changes
- Verify the fix works
### 4. Quick Commit (optional)
If not skipping commit:
```bash
git add .
git commit -m "quick: $task
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>"
```
### 5. Rubric Check (artifact tasks only)
If the task produced a new or updated work product (use case, architecture notebook, iteration plan, test plan, vision), run a quick rubric check:
- Detect artifact type from changed files
- Load the matching rubric from `.claude/rubrics/`
- Grade each criterion: ✅ satisfied / ❌ gap
- If any gaps: fix them before proceeding (quick tasks still need quality output)
Skip this step for pure code changes (bug fixes, refactors, configuration).
### 6. Quick Log (optional)
> **Scribe step** — delegate both writes to the `openup-scribe` agent (Agent
> tool, subagent_type: "openup-scribe"). You determine the values; the scribe
> only writes. Brief it with:
>
> ```
> Agent(subagent_type="openup-scribe", description="Write quick-task log entry",
> prompt="1. Append this line to docs/agent-logs/quick-tasks.log:
> [ISO timestamp] | quick-task | [task description]
> 2. [Only if task produced an artifact or decision] Append to
> .claude/memory/iteration-learnings.md:
> ## [YYYY-MM-DD] quick: [task]
> - What changed: [brief]
> - Conventions established: [any patterns worth reusing]
> Report: files written.")
> ```
After the log is written, record the gates and verify the quick-track required set. If `check-gates` exits nonzero, resolve the unmet gates before finishing:
```bash
python3 scripts/openup-state.py set-gate log_written true 2>/dev/null || true
python3 scripts/openup-state.py set-gate roadmap_synced true 2>/dev/null || true
python3 scripts/openup-state.py check-gates --require log_written,roadmap_synced 2>/dev/null || true
```
### 7. Archive State (if state was created)
```bash
python3 scripts/openup-state.py archive \
"docs/agent-logs/$(date -u +%Y)/$(date -u +%m)/$(date -u +%d)/state-quick-$(date -u +%H%M%S).json" 2>/dev/null || true
```
## Output
Returns:
- Task completed confirmation
- Files changed (count)
- Branch name (if created)
- Commit hash (if committed)
## Comparison: Standard vs Quick
| Step | Standard Workflow | Quick Task |
|------|-------------------|------------|
| Read project-status | Full document | Minimal only |
| Create branch | Task-based naming | Timestamp-based |
| SOP compliance | Full Start-of-Run | Skipped |
| Documentation | Full update | Minimal |
| Log entry | Full JSONL | Simple log line |
| **Typical time** | ~8 minutes | ~4 minutes |
## Examples
### Quick Bug Fix
```
/openup-quick-task task: "Fix typo in README.md"
```
### Documentation Update
```
/openup-quick-task task: "Update API docs for new endpoint"
```
### Skip Branching
```
/openup-quick-task task: "Add comment to utils.py" skip_branch: true
```
### Full Control
```
/openup-quick-task task: "Hot fix auth bug" skip_commit: false skip_logging: true
```
## Success Criteria
- [ ] Task completed
- [ ] Changes verified
- [ ] Branch created (if not skipped)
- [ ] Committed (if not skipped)
- [ ] Logged (if not skipped)
## Smart Features
**Auto-detect skip opportunities:**
- If already on feature branch → skip branching
- If no git changes → skip commit
- If single file change → minimal logging
**Auto-categorize task:**
- Bug fixes → `bugfix/` prefix
- Docs → `docs/` prefix
- Hot fixes → `hotfix/` prefix
## See Also
- [openup-start-iteration](../start-iteration/SKILL.md) - Full iteration workflow
- [openup-complete-task](../complete-task/SKILL.md) - Task completion with PR
- [openup-tdd-workflow](../tdd-workflow/SKILL.md) - TDD cycleRelated Skills
openup-transition
Initialize and manage Transition phase activities - deploy to users
openup-tdd-workflow
Guide Test-Driven Development cycle adapted for AI agents with a pragmatic approach
openup-sync-spec
Back-propagate pure refactors to stale artifacts; classify the diff, refuse behaviour-changes, propose targeted edits for approval (read-only by default)
openup-start-iteration
Begin a new OpenUP iteration with proper phase context and task selection
openup-shared-vision
Create shared technical vision for team alignment
openup-retrospective
Generate iteration retrospective with feedback and action items
openup-request-input
Create an input request document for asynchronous stakeholder communication
openup-readiness
Compute the change-folder dependency DAG and print READY/BLOCKED/collision report for PM intake
openup-plan-feature
Generate iteration plan and roadmap entry for a feature idea
openup-phase-review
Check phase completion criteria and prepare for phase review
openup-orchestrate
Run a full orchestrated iteration — PM decomposes the goal, delegates to specialist roles, collects outputs, and synthesizes results
openup-next
Run ONE OpenUP delivery cycle — read the derived board, claim the top READY lane, execute it, tick its progress, and exit through a legal exit. The sequential continue-loop.