implement
Skill98 starsChanged 9 months agoMITView on GitHub
Execute implementation workflow phase by phase. Use after analysis passes. Triggers on: start implementation, implement feature, begin coding.
We split this skill into the rules it gives an agent
27 rules. 0 are common, 17 are distinctive, and 10 are very specific to this project.
0
Common
Most of the 174 edited versions we read keep these.
17
Distinctive
This author's own choices. Worth a look.
10
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- could be automatic: a setting could enforce it
- Distinctive: line 43
Read progress.txt: - Check learnings from previous iterations
- Very specific: line 44
Review artifacts: - Ensure spec.md, plan.md, tasks.md, prd.json, checklist.md exist
- Very specific: line 45
Verify Analysis Passed: - Run /relentless.analyze if not already done
- Very specific: line 47
Check Branch: - Verify you're on the correct branch from PRD branchName
- Distinctive: line 53
Before marking ANY story as complete, ALL quality checks must pass.
- Very specific: line 109
If the current story has research: true in prd.json and no research file exists yet:
- Distinctive: line 118
Do NOT implement: - only research and document
- Very specific: line 128
Read prd.json to find the next story where passes: false
- Very specific: line 129
Check dependencies are met (dependent stories have passes: true)
- Distinctive: line 130
Read the story's acceptance criteria
- Distinctive: line 131
Check routing metadata (complexity, model assigned)
- Distinctive: line 140
Follow the TDD workflow above for each acceptance criterion.
- Distinctive: line 226
After completing each story, these files MUST be updated:
- Distinctive: line 250
Follow dependency order strictly
- Very specific: line 261
After completing a user story, check if ALL stories have passes: true.
- Distinctive: line 274
Skipping TDD: - Never implement without tests first
- Distinctive: line 276
Large commits: - Keep commits focused and atomic
- Distinctive: line 277
Missing typecheck: - Always run typecheck before commit
- Distinctive: line 278
Ignoring progress.txt: - Read learnings from previous iterations
- Very specific: line 279
Not checking queue: - Always check .queue.txt for user input
- Very specific: line 280
Skipping analysis: - Run /relentless.analyze before implementing
- Distinctive: line 281
Ignoring routing metadata: - Check story complexity and model assignment
- Distinctive: line 289
Never skip TDD - tests come FIRST
- Distinctive: line 290
Never skip quality checks
- Distinctive: line 291
Commit after each story
- Distinctive: line 292
Update ALL tracking files
- Very specific: line 293
Check .queue.txt for mid-run input
“Common” is measured: kept by at least 6 in 10 of the 174 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Implementation Workflow Executor
Guide systematic implementation of features using TDD and quality-first approach.
Note: This skill uses generic placeholders. Adapt commands and paths to your project:
- Quality commands: Check your
package.json,Makefile, or build config- File paths: Adjust based on your project structure
- Examples show common conventions - your project may differ
SpecKit Workflow
This skill is Step 6 of 6 in the Relentless workflow:
specify → plan → tasks → convert → analyze → implement
Prerequisites:
spec.md- Feature specificationplan.md- Technical implementation plantasks.md- User stories with acceptance criteriaprd.json- Converted PRD with routing metadatachecklist.md- Quality validation checklist- Analysis must PASS (run
/relentless.analyzefirst)
The Job
Implement user stories one at a time, following strict TDD and updating all tracking files.
Before You Start
- Read the Constitution - Review your project's constitution (if exists) for governance principles
- Read prompt.md - Review workflow guidelines (if exists)
- Read progress.txt - Check learnings from previous iterations
- Review artifacts - Ensure spec.md, plan.md, tasks.md, prd.json, checklist.md exist
- Verify Analysis Passed - Run
/relentless.analyzeif not already done - Identify Quality Commands - Find your project's typecheck, lint, and test commands
- Check Branch - Verify you're on the correct branch from PRD
branchName
CRITICAL: Quality Gates (Non-Negotiable)
Before marking ANY story as complete, ALL quality checks must pass.
Find your project's commands (examples for different ecosystems):
# JavaScript/TypeScript (npm/yarn/bun/pnpm):
npm run typecheck && npm run lint && npm test
# Python:
mypy . && ruff check . && pytest
# Go:
go build ./... && golangci-lint run && go test ./...
# Rust:
cargo check && cargo clippy && cargo testRequirements:
- Typecheck: 0 errors
- Lint: 0 errors AND 0 warnings
- Tests: All pass
If ANY check fails, DO NOT mark the story as complete. Fix the issues first.
TDD Workflow (MANDATORY)
For EVERY story, follow strict Test-Driven Development:
Step 1: Write Failing Tests First (RED)
# Create test file if needed
# Write tests that define expected behavior
# Run your test command - tests MUST fail initiallyStep 2: Implement Minimum Code (GREEN)
# Write only enough code to pass tests
# Run tests - they MUST pass nowStep 3: Refactor
# Clean up while keeping tests green
# Run tests - they MUST still passDo NOT skip TDD. Tests are contracts that validate your implementation.
Research Phase (if story has research: true)
If the current story has research: true in prd.json and no research file exists yet:
- Explore the codebase - Find relevant files, patterns, and dependencies
- Document findings in
relentless/features/<feature>/research/<story-id>.md:- Existing patterns that should be followed
- Files that will likely need modification
- Dependencies and integration points
- Potential gotchas or edge cases
- Recommended implementation approach
- Do NOT implement - only research and document
- Save your findings to the research file and end your turn
Per-Story Implementation Flow
For each story (in dependency order):
1. Identify the Story
- Read
prd.jsonto find the next story wherepasses: false - Check dependencies are met (dependent stories have
passes: true) - Read the story's acceptance criteria
- Check routing metadata (complexity, model assigned)
2. Find Relevant Checklist Items
- Open
checklist.md - Find items tagged with
[US-XXX]for this story - Note any governance/compliance items
- Ensure Quality Gates and TDD items are included
3. Implement with TDD
Follow the TDD workflow above for each acceptance criterion.
4. Update tasks.md
As you complete each criterion:
# Change from:
- [ ] Criterion text
# To:
- [x] Criterion text5. Update checklist.md
For each verified checklist item:
# Change from:
- [ ] CHK-XXX [US-001] Description
# To:
- [x] CHK-XXX [US-001] Description6. Run Quality Checks
# Run your project's quality commands
# All must pass with 0 errors/warnings7. Commit Changes
git add -A
git commit -m "feat: US-XXX - Story Title"8. Update prd.json
Set the story's passes field to true:
{
"id": "US-001",
"title": "...",
"passes": true, // <- Change from false to true
...
}9. Update progress.txt
Append progress entry:
## [Date] - US-XXX: Story Title
**Implemented:**
- What was built
- Key decisions made
**Files Changed:**
- path/to/file (new/modified)
**Tests Added:**
- path/to/test.file
**Learnings:**
- Patterns discovered
- Gotchas encountered
**Constitution Compliance:**
- [list principles followed]
---Check for Queued Prompts
Between iterations, check .queue.txt for user input:
# If .queue.txt exists, read and process it
# Acknowledge in progress.txt
# Process in FIFO orderFile Update Summary
After completing each story, these files MUST be updated:
| File | Update |
tasks.md | Check off - [x] completed acceptance criteria |
checklist.md | Check off - [x] verified checklist items |
prd.json | Set "passes": true for the story |
progress.txt | Append progress entry with learnings |
Implementation Phases
Phase 0: Setup
- Infrastructure, tooling, configuration
- Usually US-001 type stories
Phase 1: Foundation
- Data models, types, schemas
- Base utilities and helpers
- Core infrastructure
Phase 2: User Stories
- Feature implementation
- Follow dependency order strictly
Phase 3: Polish
- E2E tests
- Documentation
- Performance optimization
Stop Condition
After completing a user story, check if ALL stories have passes: true.
If ALL stories are complete and passing, output:
<promise>COMPLETE</promise>If there are still stories with passes: false, end your response normally (another iteration will pick up the next story).
Common Pitfalls to Avoid
- Skipping TDD - Never implement without tests first
- Suppressing lints - Fix issues properly, don't disable rules
- Large commits - Keep commits focused and atomic
- Missing typecheck - Always run typecheck before commit
- Ignoring progress.txt - Read learnings from previous iterations
- Not checking queue - Always check
.queue.txtfor user input - Skipping analysis - Run
/relentless.analyzebefore implementing - Ignoring routing metadata - Check story complexity and model assignment
Notes
- Work on ONE story at a time
- Follow dependency order strictly
- Never skip TDD - tests come FIRST
- Never skip quality checks
- Commit after each story
- Update ALL tracking files
- Check
.queue.txtfor mid-run input - This is a guided workflow for systematic implementation
- Adapt all commands and paths to your project's specific setup
Other versions191 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 191
- mattpocock/skills★ 303ksame name
- QuivrHQ/quivr★ 40ksame name
- foambubble/foam★ 17ksame name
- formbricks/formbricks★ 13ksame name
- ArcReel/ArcReel★ 6ksame name
- vinvcn/mattpocock-skills-zh-CN★ 5.3ksame name
- fallow-rs/fallow★ 5.1ksame name
- withcoral/coral★ 4.9ksame name
- bestofjs/bestofjs★ 3.1ksame name
- huggingface/OpenEnv★ 2.8ksame name
181 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
Used alongsideProjects with it also use setup-ts-deep-modules, ask-matt and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
liufeng1976Skillsetup-ts-deep-modules
Wire dependency-cruiser into a TypeScript repo so each package is a deep module — implementation hidden in subfolders, reachable only through its entry-point files. User-invoked.
Also in 33 projects that use this
GoogleSkillask-matt
Ask which skill or flow fits your situation. A router over the skills in this repo.
Also in 79 projects that use this
AWSSkillto-questionnaire
Turn a decision you can't fully answer into a questionnaire for someone else to fill in.
Also in 76 projects that use this
GoogleSkillto-tickets
Break a plan, spec, or the current conversation into a set of tracer-bullet tickets, each declaring its blocking edges, published to the configured tracker — edges as text in one file per ticket locally, or native blocking links on a real tracker.
Also in 138 projects that use this
RoentekSkillloop-me
Grill me about specs for the workflows I want to build, within this workspace.
Also in 42 projects that use this
ryotaroh180105Skillclaude-handoff
Hand the current conversation off to a fresh background agent that picks up the work immediately.
Also in 34 projects that use this
- +2 more from google
In this repository15 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- analyzeSkill
- checklistSkill
- clarifySkill
- constitutionSkill
- convertSkill
- learnSkill
- planSkill
- prdSkill
- relentlessSkill
- specifySkill
- tasksSkill
- taskstoissuesSkill
- validatorsSkill
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

