implement-task
Run approved task files through builder agents in isolated worktrees, up to 3 in parallel, test-first.
Skill by YIlyaA
Chalk Labs
Need help deploying implement-task?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/yilyaa-implement-task
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: testing, productivity
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md and code. Checked Nov 1, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/implement-task
What's inside
- Skill name
implement-task- Description
- Run approved task files through builder agents in isolated worktrees, up to 3 in parallel, test-first. Use when one or more tasks in tasks/ have status approved.
- License
- MIT
- Instructions
- 27 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Implement Task Skill ## Purpose Turn approved tasks into verified branches as fast as possible without agents stepping on each other. ## Agent Instructions ### Step 1: Pick runnable tasks List `tasks/T-*.md` with `status: approved` whose `blocked_by` tasks are all `done`. Take at most 3 with `parallel_ok: true`; a task with `parallel_ok: false` runs alone. ### Step 2: Start builders Set each picked task to `status: in-progress` and record the start time. Spawn one `builder` agent per task in the same message, so they run concurrently (each gets its own worktree from `isolation: worktree`). Pass only the task file path. ### Step 3: Collect results For each builder: check it reports `uv run python scripts/check.py` output ending in `ALL CHECKS PASSED`. If not, or if it stopped on budget or diff limit, mark the task `in-progress` with the blocker noted and report it. Do not retry silently. ### Step 4: Hand off to verification For each green branch, run the `verify-diff` skill. ## Expected Output `| id | branch | checks | minutes | stopped early? |`, then what goes to verification. ## Guardrails - Only `approved` tasks. Max 3 parallel builders (8 GB RAM). - Builders never merge, never push to `main`, never touch guardrail files.
From YIlyaA/agent-workbench, MIT.
More from YIlyaA/agent-workbench
- 0.566score
bootstrap
YIlyaA
Set up the agent work environment in a new or existing project (AGENTS.md, CONTEXT.md, skills, planner/builder/verifier agents, guardrails, metrics, readiness…
SkillAFreshness: AClaude Code +9 - 0.565score
plan-task
YIlyaA
Turn a goal for Agents Workbench into one or more task files in tasks/ with a frozen est_hours_old_way, then stop for approval.
SkillAFreshness: AClaude Code +9 - 0.565score
verify-diff
YIlyaA
Independent review of a builder's branch by the verifier agent in a fresh context; records the verdict and finding count in the task file.
SkillAFreshness: AClaude Code +9 - 0.560score
release
YIlyaA
Open the PR for a verified task branch, and after the human merges it, log the tasks.csv row, run the metrics report and update release evidence.
SkillAFreshness: AClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Oct 16, 2026 (20 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/YIlyaA/agent-workbench
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from YIlyaA
- 0.566score
bootstrap
YIlyaA
Set up the agent work environment in a new or existing project (AGENTS.md, CONTEXT.md, skills, planner/builder/verifier agents, guardrails, metrics, readiness…
SkillAFreshness: AClaude Code +9 - 0.565score
plan-task
YIlyaA
Turn a goal for Agents Workbench into one or more task files in tasks/ with a frozen est_hours_old_way, then stop for approval.
SkillAFreshness: AClaude Code +9 - 0.565score
verify-diff
YIlyaA
Independent review of a builder's branch by the verifier agent in a fresh context; records the verdict and finding count in the task file.
SkillAFreshness: AClaude Code +9 - 0.560score
release
YIlyaA
Open the PR for a verified task branch, and after the human merges it, log the tasks.csv row, run the metrics report and update release evidence.
SkillAFreshness: AClaude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get implement-task running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
implement-task skill. Agent Mart. https://agentmart.dev/skills/yilyaa-implement-task (last updated 2026-10-24)