Automated summary
Dispatches tasks to an orchestrator for autonomous parallel execution with built-in verification.
Strengths
- Automates batch refactoring
- Runs tasks independently in isolated worktrees
- Verifies changes before commit
- Allows background execution
Limitations
- Requires clear task descriptions
- May not handle interdependent tasks well
- Can be overkill for simple quick fixes
Use for multi-step tasks that can run independently and benefit from verification.
Avoid for single quick fixes or exploratory debugging.
Security analysis
SafeThe skill describes a task orchestrator CLI for batch code changes with isolation and verification. It does not instruct destructive commands, exfiltration, or disabling safety; it merely documents how to use the tool safely. The allowed-tools restriction to Bash(undercity:*) limits execution to undercity commands, which themselves operate in isolated worktrees.
No concerns found
How this analysis is madeExamples
Fix all TypeScript strict null check errors in src/Add JSDoc comments to all exported functionsMigrate deprecated API calls to v2name: undercity description: Dispatch tasks to the undercity orchestrator for autonomous batch execution. Use when adding tasks, checking status, or running parallel work. Ideal for multi-task workloads that benefit from autonomous verification. allowed-tools: Bash(undercity:*)
Undercity Task Orchestrator
Dispatch work to undercity for autonomous, parallel execution with built-in verification (typecheck, test, lint).
When to Use Undercity
Good fit:
- Multi-step tasks that can run independently
- Batch refactoring across many files
- Work that benefits from verification loops
- Tasks to run overnight/in background
Skip undercity for:
- Single quick fixes (do directly)
- Questions/explanations
- Exploratory debugging
Commands
Add a Task
undercity add "task description"
Examples:
undercity add "Fix all TypeScript strict null check errors in src/"
undercity add "Add JSDoc comments to all exported functions"
undercity add "Migrate deprecated API calls to v2"
Add Task with Context (Handoff)
Pass context to help workers start with relevant information:
# Pass files you've already analyzed
undercity add "Refactor auth module" --files-read "src/auth.ts,src/types.ts"
# Pass notes about decisions or constraints
undercity add "Fix validation bug" --notes "Issue is in validateInput(), not the schema"
Check Task Board
# All tasks
undercity tasks
# Only pending
undercity tasks --status pending
# Only completed
undercity tasks --status complete
Mark Tasks Complete
# Mark task as complete
undercity complete <task-id>
# With resolution notes
undercity complete <task-id> --resolution "Fixed in commit abc"
# With reason (for closures without implementation)
undercity complete <task-id> --reason "Already implemented"
Start Autonomous Execution
# Process all pending tasks
undercity grind
# Limit concurrent workers
undercity grind --parallel 2
# Process specific number of tasks
undercity grind -n 5
Proactive PM (Task Generation)
# Generate tasks from codebase analysis
undercity pm --propose
# Research a topic via web search
undercity pm "topic" --research
# Full ideation: research + propose
undercity pm "topic" --ideate
# Add generated tasks to board
undercity pm "topic" --ideate --add
Monitor Progress
# Live TUI dashboard
undercity watch
# Current grind session status (JSON)
undercity status
Task Description Guidelines
Write clear, actionable task descriptions:
Good:
- "Fix TypeScript error TS2345 in src/api/client.ts"
- "Add input validation to all POST endpoints in src/routes/"
- "Replace deprecated lodash _.pluck with _.map"
Vague (avoid):
- "Fix bugs" (which bugs?)
- "Improve performance" (where? how?)
- "Clean up code" (too broad)
How It Works
- Tasks execute in isolated git worktrees (no conflicts)
- All changes verified (typecheck, test, lint) before commit
- Tasks merge to main serially
- Failed tasks retry with model escalation (sonnet -> opus)
Task Board Location
Tasks are stored in SQLite at .undercity/undercity.db (gitignored).
Use CLI commands (undercity tasks, undercity add, etc.) to manage tasks.
Runtime state files (knowledge, metrics, recovery) are in .undercity/ and gitignored.
From the file to a first result
The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.
View the source and supporting filesThis download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.
1. Download the file
Download SKILL.mdFrom the root of a test project, save the file as .skill-downloads/undercity-task-orchestrator-mmbizjdv/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.
Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.
mkdir -p '.skill-downloads/undercity-task-orchestrator-mmbizjdv' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/undercity-task-orchestrator-mmbizjdv/download' --output '.skill-downloads/undercity-task-orchestrator-mmbizjdv/SKILL.md.part' && mv '.skill-downloads/undercity-task-orchestrator-mmbizjdv/SKILL.md.part' '.skill-downloads/undercity-task-orchestrator-mmbizjdv/SKILL.md'2. Check prerequisites
Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.
Tools declared by the author (check availability in your agent): Bash(undercity:*)
No recognized prerequisites section in this file. This does not mean it is self-contained: libraries, tools and access requirements may be mentioned elsewhere in the content or source repository.
The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.
3. Load the skill
After inspection, copy the complete folder into your project using one of the commands below, then open a new tool session. The command stops if a folder with the same name already exists: compare the files before replacing it.
Claude Code
test -f '.skill-downloads/undercity-task-orchestrator-mmbizjdv/SKILL.md' && test -s '.skill-downloads/undercity-task-orchestrator-mmbizjdv/SKILL.md' && mkdir -p '.claude/skills' && test ! -e '.claude/skills/undercity' && cp -R '.skill-downloads/undercity-task-orchestrator-mmbizjdv' '.claude/skills/undercity'Cursor
test -f '.skill-downloads/undercity-task-orchestrator-mmbizjdv/SKILL.md' && test -s '.skill-downloads/undercity-task-orchestrator-mmbizjdv/SKILL.md' && mkdir -p '.cursor/skills' && test ! -e '.cursor/skills/undercity' && cp -R '.skill-downloads/undercity-task-orchestrator-mmbizjdv' '.cursor/skills/undercity'OpenSkills
OpenSkills must already be installed. The local install/read procedure is checked with OpenSkills 1.5.0, which requires Node.js ≥ 20.6.0. Catalog slugs are not resolved automatically.
Official OpenSkills installationopenskills install './.skill-downloads/undercity-task-orchestrator-mmbizjdv'openskills read 'undercity-task-orchestrator-mmbizjdv'Reading should display the installed file content and its base directory. It does not run the skill or install its dependencies.
For an agent that reads AGENTS.md, then sync the index. This command changes AGENTS.md: review the diff before using it.
openskills syncManual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.
4. Try a first task
Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:
Read .skill-downloads/undercity-task-orchestrator-mmbizjdv/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.
React Hook Form Best Practices
Development
Master React Hook Form: avoid anti-patterns, manage subscriptions, and handle reset, dirty state, and controlled inputs correctly.
Emotion to Mantine Migration
Development
Migrate Emotion styled-components to Mantine components with style props and CSS modules. Use when converting .styled.tsx files or removing @emotion imports.
Scope Creep Detector
Development
Analyzes git diffs against a stated intent to detect scope creep, unrelated files, broad PRs, and oversized changes, then suggests keep, split, or justify actions.