Subagent-Driven Development
HoangNguyen0403/agent-skills-standard
Runs a multi-task implementation plan by sending each task to a fresh implementer subagent, reviewing it independently, then reviewing the whole branch.
Agent skill
Implementing Tasks with Subagent Review
by prime-radiant-inc in prime-radiant-inc/iterative-development
Runs a batch of TDD-sized tasks by dispatching an implementer subagent per task, followed by two-stage parallel adversarial review and fix loops.
Install Implementing Tasks with Subagent Review
$ npx skills add prime-radiant-inc/iterative-development --skill implementing-tasks -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install prime-radiant-inc/iterative-development implementing-tasks --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/prime-radiant-inc/iterative-development.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/implementing-tasks .claude/skills/implementing-tasks && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "implementing-tasks" agent skill from https://github.com/prime-radiant-inc/iterative-development/tree/main/skills/implementing-tasks into .claude/skills/implementing-tasks/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "implementing-tasks", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/prime-radiant-inc/iterative-development/tree/main/skills/implementing-tasksType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add prime-radiant-inc/iterative-development --skill implementing-tasks -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install prime-radiant-inc/iterative-development implementing-tasks --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/prime-radiant-inc/iterative-development.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/implementing-tasks .agents/skills/implementing-tasks && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "implementing-tasks" agent skill from https://github.com/prime-radiant-inc/iterative-development/tree/main/skills/implementing-tasks into .agents/skills/implementing-tasks/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "implementing-tasks", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add prime-radiant-inc/iterative-development --skill implementing-tasks -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install prime-radiant-inc/iterative-development implementing-tasks --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/prime-radiant-inc/iterative-development.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/implementing-tasks .cursor/skills/implementing-tasks && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "implementing-tasks" agent skill from https://github.com/prime-radiant-inc/iterative-development/tree/main/skills/implementing-tasks into .cursor/skills/implementing-tasks/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "implementing-tasks", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/prime-radiant-inc/iterative-development.git --path skills/implementing-tasks--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add prime-radiant-inc/iterative-development --skill implementing-tasks -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install prime-radiant-inc/iterative-development implementing-tasks --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/prime-radiant-inc/iterative-development.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/implementing-tasks .gemini/skills/implementing-tasks && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "implementing-tasks" agent skill from https://github.com/prime-radiant-inc/iterative-development/tree/main/skills/implementing-tasks into .gemini/skills/implementing-tasks/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "implementing-tasks", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install prime-radiant-inc/iterative-development implementing-tasksInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add prime-radiant-inc/iterative-development --skill implementing-tasks -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/prime-radiant-inc/iterative-development.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/implementing-tasks .github/skills/implementing-tasks && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "implementing-tasks" agent skill from https://github.com/prime-radiant-inc/iterative-development/tree/main/skills/implementing-tasks into .github/skills/implementing-tasks/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "implementing-tasks", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add prime-radiant-inc/iterative-development --skill implementing-tasks -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install prime-radiant-inc/iterative-development implementing-tasks --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/prime-radiant-inc/iterative-development.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/implementing-tasks .opencode/skills/implementing-tasks && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "implementing-tasks" agent skill from https://github.com/prime-radiant-inc/iterative-development/tree/main/skills/implementing-tasks into .opencode/skills/implementing-tasks/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "implementing-tasks", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
implementing-tasks- GitHub stars
- 181
- Token cost
- ~1.2k tokens
- SKILL.md length
- 551 words
- Files
- 4
- Skills in repo
- 6
- Repo updated
- First seen
- Licence
- Apache-2.0
At a glance
Runs a batch of TDD-sized tasks by dispatching an implementer subagent per task, followed by two-stage parallel adversarial review and fix loops.
- Works in 5 steps: Dispatch implementer → Handle implementer status → PAR spec-compliance review (Stage 1) → …
- Executing a batch of small test-driven tasks inside an iterative development loop
- SKILL.md covers Overview, When to Use, Per-Task Cycle and Model Selection, plus 3 more sections
- Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
What it does
Called from a `running-an-iteration` skill with an in-memory task list, this skill takes each task through a fixed cycle: an implementer subagent working test-first, a spec-compliance review, a fix loop, a code-quality review that includes a boxing-in check, another fix loop, and finally marking the task complete. It is a fork of a subagent-driven development skill with the plan-file reading phase and the final end-of-plan reviewer removed.
The implementer must first complete a pre-flight mapping from acceptance criteria to proof seam to scenario, and is re-dispatched if it skips it. Its status then decides what happens: DONE moves to review, DONE_WITH_CONCERNS means reading the concerns, NEEDS_CONTEXT gets the missing context, and BLOCKED is handled by adding context, using a more capable model, splitting the task or escalating. Each review stage sends two reviewers in parallel under competitive framing, merges findings by union with the worst severity, and loops until the stage passes. Prompt templates for the implementer and both reviewers are included.
When your agent uses it
- Executing a batch of small test-driven tasks inside an iterative development loop
- Running implementation with separate spec-compliance and code-quality reviews
- Handling implementer subagent statuses such as blocked or needs context
Example prompts
- “Implement these tasks one by one with an implementer subagent and two-stage review.”
- “Work through the sprint's task list test-first and report each task's completion status.”
- “The implementer reported BLOCKED on the caching task, so re-dispatch it with more context.”
Requirements
- The running-an-iteration skill from the iterative development plugin, which supplies the tasks
- An agent that can dispatch subagents
Workflow steps
5 steps, taken from the step headings in SKILL.md.
What it can do on your machine
Read from SKILL.md and the folder at commit c05889a. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
No scripts in the folder and no shell commands in SKILL.md.
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Implementing Tasks with Subagent Review loads about 1.2k tokens when it runs. Until then it costs about 56 tokens; SKILL.md has 551 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
The full file from prime-radiant-inc/iterative-development at commit c05889a, republished under its Apache-2.0 licence (© prime-radiant-inc). 551 words, ~1,208 tokens.
Download SKILL.mdSave it as
.claude/skills/implementing-tasks/SKILL.md (or your agent's skills folder). This skill also uses 3 other files; get the full folder from GitHub.- name
- implementing-tasks
- description
- Use when executing a batch of TDD-sized tasks inside a running-an-iteration call — dispatches an implementer subagent per task following red-green-refactor discipline and returns per-task completion status.
Implementing Tasks
Overview
Takes an in-memory batch of TDD-sized tasks and executes each through: implementer subagent (TDD) → PAR spec-compliance review → fix loop → PAR code-quality review with boxing-in check → fix loop → mark complete. This is a fork of superpowers:subagent-driven-development with the plan-file reading phase stripped and the final end-of-plan reviewer removed.
When to Use
Invoked by running-an-iteration with a list of tasks. Tasks are passed in memory, not via a file.
Per-Task Cycle
For each task in the provided list:
1. Dispatch implementer
Using the template in implementer-subagent-prompt.md, dispatch a single implementer subagent with:
- The full task description and context
- The proof obligations for each observable AC in the task's stories
- The list of existing scenarios that may be impacted
The implementer MUST complete a pre-flight mapping (AC → proof seam → scenario) before writing code. If the implementer skips the pre-flight, re-dispatch with explicit instructions to complete it first.
2. Handle implementer status
- DONE: proceed to spec-compliance review (step 3). Verify the implementer's report includes pre-flight mapping and scenario updates.
- DONE_WITH_CONCERNS: read the concerns. If about correctness/scope, address before review. If observations, note and proceed.
- NEEDS_CONTEXT: provide the missing context and re-dispatch
- BLOCKED: assess: context problem → re-dispatch with context; too hard → re-dispatch with more capable model; task too large → break into smaller pieces; plan wrong → escalate to caller
3. PAR spec-compliance review (Stage 1)
Following skills/shared/parallel-adversarial-review.md:
- Build spec-compliance prompt using
spec-compliance-reviewer-prompt.md- Include the proof obligations and the implementer's evidence claims
- Wrap in PAR competitive framing from
skills/shared/par-reviewer-wrapper.md - Dispatch TWO spec-compliance reviewers in parallel
- Aggregate findings (PAR rules: union of findings, severity = take worst)
- If ❌ issues found:
- Send aggregated issues back to the implementer subagent (same subagent, via continuation message)
- Implementer fixes
- Re-dispatch fresh PAR spec-compliance pair
- Repeat until ✅ spec compliant with adequate evidence
- Only proceed to Stage 2 after Stage 1 is ✅
Show full SKILL.md (245 more words)Show less
4. PAR code-quality review (Stage 2)
Following skills/shared/parallel-adversarial-review.md:
- Build code-quality prompt using
code-quality-reviewer-prompt.md- Include the next 3 pending roadmap iterations for the boxing-in check
- Include the implementer's corpus contribution for quality review
- Wrap in PAR competitive framing
- Dispatch TWO code-quality reviewers in parallel
- Aggregate findings
- If ❌ changes needed:
- Send aggregated issues back to the implementer
- Implementer fixes
- Re-dispatch fresh PAR code-quality pair
- Repeat until ✅ approved
5. Mark task complete
Record the task as done. Move to the next task.
After all tasks complete, return a per-task result list to the caller, including:
- Per-task status
- Scenarios added or updated per task
- Evidence commands per task
Model Selection
Use the least powerful model that can handle each role:
| Role | Signal → Model |
|---|---|
| Implementer (mechanical: 1-2 files, clear spec) | Cheap/fast model |
| Implementer (integration: multi-file, judgment) | Standard model |
| Spec-compliance reviewer | Standard model |
| Code-quality reviewer | Most capable model |
Quick Reference
| Per task | Subagents dispatched |
|---|---|
| Implementer | 1 (sequential, TDD) |
| Spec-compliance review (PAR) | 2 in parallel |
| Code-quality review (PAR) | 2 in parallel |
| Minimum per task | 5 (before re-review loops) |
Red Flags
- Never start code-quality review before spec compliance is ✅
- Never skip the re-review after fixes (reviewer found issues = implementer fixes = review again)
- Never dispatch multiple implementers in parallel (conflicts)
- Never accept "close enough" on spec compliance
- Never let implementer self-review replace the two-stage review
References
implementer-subagent-prompt.md— implementer dispatch templatespec-compliance-reviewer-prompt.md— Stage 1 review templatecode-quality-reviewer-prompt.md— Stage 2 review template (includes boxing-in)skills/shared/parallel-adversarial-review.md— PAR methodologyskills/shared/par-reviewer-wrapper.md— competitive framing wrapper
© prime-radiant-inc, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
SKILL.md and 3 other files in skills/implementing-tasks of prime-radiant-inc/iterative-development.
- SKILL.md
- code-quality-reviewer-prompt.md
- implementer-subagent-prompt.md
- spec-compliance-reviewer-prompt.md
Open the folder on GitHubat commit c05889a
Compare with similar skills
Implementing Tasks with Subagent Review next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Implementing Tasks with Subagent Review this skillprime-radiant-inc/iterative-development | 181 | — | ~1.2k | Automated safety check: Pass | Apache-2.0 | |
| Subagent-Driven DevelopmentHoangNguyen0403/agent-skills-standard | 571 | — | ~1.3k | Automated safety check: Pass | MIT | |
| Loop Change Verifiercobusgreyling/loop-engineering | 11k | 1 repos | ~383 | Automated safety check: Pass | MIT | |
| Writing Skillsed3dai/ed3d-plugins | 250 | 1 repos | ~1.3k | Automated safety check: Pass | None | |
| Assign To Workforceagentculture/culture | 113 | — | ~2.7k | Automated safety check: Warn | Apache-2.0 | |
| Ralph Wiggum V2majiayu000/claude-skill-registry | 666 | 1 repos | ~1.9k | Automated safety check: Pass | MIT |
Similar skills
Loop Change Verifier
cobusgreyling/loop-engineering
Acts as a skeptical checker for changes an implementer sub-agent made, running the tests, confirming the diff scope and returning approve, reject or escalate to a human.
Writing Skills
ed3dai/ed3d-plugins
A skill your agent uses when creating new skills, editing existing skills, or verifying skills work before deployment - applies TDD to process documentation by testing with subagents before writing…
Assign To Workforce
agentculture/culture
Fan out a converged devague plan's dependency waves to parallel agents in isolated git worktrees, one agent per task per wave, with TDD-gated merges by the main agent.
Ralph Wiggum V2
majiayu000/claude-skill-registry
Autonomous TDD development loop with parallel agent swarm, category evolution, and convergence detection.
Writing Skills
sandgardenhq/sgai
TDD for process documentation - test with subagents before writing, iterate until bulletproof.
More from prime-radiant-inc/iterative-development
Extracting Requirements
prime-radiant-inc/iterative-development
Reads human-written spec documents and produces per-epic requirement files with proof obligations plus behavior scenarios with stable IDs, using parallel chunked extraction.
Walking Skeleton Roadmap Scoping
prime-radiant-inc/iterative-development
Turns extracted requirements into a roadmap by choosing a walking skeleton iteration with its first journey scenario and ordering the remaining work into follow-on iterations.
Iterative Development Orchestrator
prime-radiant-inc/iterative-development
Runs an autonomous loop that extracts requirements with proof obligations, builds a walking skeleton, then audits sprint by sprint against real behavior evidence.
Running An Iteration
prime-radiant-inc/iterative-development
A skill your agent uses when executing the next pending iteration from an iterative-development roadmap — picks the iteration, decomposes into code and evidence tasks, runs sentinel corpus baseline…
Iteration Progress Audit
prime-radiant-inc/iterative-development
Checks the quality of behavior evidence after each iteration in three tiers, using two auditor subagents in parallel to review the same work and find gaps.
Related
Categories
Questions about Implementing Tasks with Subagent Review
What does Implementing Tasks with Subagent Review do?
Runs a batch of TDD-sized tasks by dispatching an implementer subagent per task, followed by two-stage parallel adversarial review and fix loops. Called from a `running-an-iteration` skill with an in-memory task list, this skill takes each task through a fixed cycle: an implementer subagent working test-first, a spec-compliance review, a fix loop, a code-quality review that includes a boxing-in check, another fix loop, and finally marking the task complete. It is a fork of a subagent-driven development skill with the plan-file reading phase and the final end-of-plan reviewer removed.
When should I use Implementing Tasks with Subagent Review?
Implementing Tasks with Subagent Review fits situations like: executing a batch of small test-driven tasks inside an iterative development loop; running implementation with separate spec-compliance and code-quality reviews; handling implementer subagent statuses such as blocked or needs context.
How do I install Implementing Tasks with Subagent Review in Claude Code?
Run `npx skills add prime-radiant-inc/iterative-development --skill implementing-tasks -a claude-code`. Or copy the skill folder (skills/implementing-tasks in prime-radiant-inc/iterative-development) into .claude/skills/implementing-tasks in your project. Claude Code loads it when a task matches its description.
How do I install Implementing Tasks with Subagent Review in Codex?
Run `npx skills add prime-radiant-inc/iterative-development --skill implementing-tasks -a codex`. Or copy the skill folder (skills/implementing-tasks in prime-radiant-inc/iterative-development) into .agents/skills/implementing-tasks in your project. Codex loads it when a task matches its description.
Can I use Implementing Tasks with Subagent Review in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add prime-radiant-inc/iterative-development --skill implementing-tasks -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/implementing-tasks, .gemini/skills/implementing-tasks, .github/skills/implementing-tasks and .opencode/skills/implementing-tasks in your project.
What does Implementing Tasks with Subagent Review need to run?
SKILL.md names no scripts, command-line tools or credentials: Implementing Tasks with Subagent Review is instructions for the agent only. Our summary lists: The running-an-iteration skill from the iterative development plugin, which supplies the tasks; An agent that can dispatch subagents.
Does Implementing Tasks with Subagent Review access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Implementing Tasks with Subagent Review safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Implementing Tasks with Subagent Review use?
Implementing Tasks with Subagent Review is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Implementing Tasks with Subagent Review use?
About 1.2k tokens (SKILL.md is roughly 4.8k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Implementing Tasks with Subagent Review?
Skills that share tags, products or a category with Implementing Tasks with Subagent Review: Subagent-Driven Development (HoangNguyen0403/agent-skills-standard, 571 stars), Loop Change Verifier (cobusgreyling/loop-engineering, 11k stars), Writing Skills (ed3dai/ed3d-plugins, 250 stars) and Assign To Workforce (agentculture/culture, 113 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Implementing Tasks with Subagent Review?
prime-radiant-inc (a GitHub organization) maintains it in prime-radiant-inc/iterative-development, which has 181 GitHub stars. The repository holds 6 skills in this directory. The repository was last updated on June 6, 2026.
Source: prime-radiant-inc/iterative-development on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.