agentleFS
SireJeffSireJeff/k0ntext/skills/implement/SKILL.md

implement

Skill8 starsChanged 9 months agoMITView on GitHub

RPI Implement Phase - Execute chunk-based todolists with atomic changes and continuous testing

191 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects 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.
  • liufeng1976Skill

    setup-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 43 projects that use this

  • GoogleSkill

    ask-matt

    Ask which skill or flow fits your situation. A router over the skills in this repo.

    Also in 91 projects that use this

  • RoentekSkill

    loop-me

    Grill me about specs for the workflows I want to build, within this workspace.

    Also in 50 projects that use this

  • AWSSkill

    to-questionnaire

    Turn a decision you can't fully answer into a questionnaire for someone else to fill in.

    Also in 88 projects that use this

  • ryotaroh180105Skill

    claude-handoff

    Hand the current conversation off to a fresh background agent that picks up the work immediately.

    Also in 42 projects that use this

  • GoogleSkill

    to-spec

    Turn the current conversation into a spec and publish it to the project issue tracker — no interview, just synthesis of what you've already discussed.

    Also in 168 projects that use this

  • +3 more from google

The file

Show its rules and text

We split this skill into the rules it gives an agent

16 rules. 0 are common, 13 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 172 edited versions we read keep these.

  • 13

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    When invoked, execute the approved implementation plan chunk by chunk:

    line 7
  2. Very specific:

    If not found, run /context-eng:plan $ARGUMENTS first

    line 21
  3. Distinctive:

    Verify plan status is APPROVED

    line 62
  4. Distinctive:

    If pass: commit with message

    line 77
  5. Distinctive:

    If fail: STOP, investigate, fix

    line 78
  6. Distinctive:

    Update CHUNK-Rn in research to IMPLEMENTED

    line 82
  7. Distinctive:

    Commit chunk documentation updates

    line 83
  8. Distinctive:

    After every 3 chunks: reload plan

    line 86
  9. Distinctive:

    If >35% utilization: save, compact, continue

    line 87
  10. Distinctive:

    After all chunks complete

    line 90
  11. Very specific:

    Check CODE_TO_WORKFLOW_MAP.md for affected workflows

    line 93
  12. Distinctive:

    Update workflow files with new line numbers

    line 94
  13. Distinctive:

    Update function signatures if changed

    line 95
  14. Distinctive:

    Update chunk progress in plan

    line 98
  15. Very specific:

    Run /context-eng:validate to verify

    line 106
  16. Distinctive:

    Test results (per chunk): ~5k tokens

    line 149

“Common” is measured: kept by at least 6 in 10 of the 172 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Context Engineering: Implement Phase (Enhanced)

When invoked, execute the approved implementation plan chunk by chunk:

Key Innovation: Inter-Phase Awareness

This implement phase KNOWS:

  • RPI-Plan structured chunks for atomic implementation
  • Each CHUNK-Pn contains a complete, ordered todolist
  • Chunk dependencies dictate execution order
  • Marking chunks complete updates both plan AND research documents
  • Context reset is needed after every 3 chunks or 35% utilization

Prerequisites

  • Approved plan at .claude/plans/active/[feature]_plan.md
  • Plan contains chunk manifest with chunk-todolists
  • If not found, run /context-eng:plan $ARGUMENTS first

Golden Rules

ONE CHUNK → COMPLETE TODOLIST → MARK DONE → NEXT CHUNK
ONE TODO → ONE CHANGE → ONE TEST → ONE COMMIT

Chunk-Based Implementation Loop

FOR each CHUNK-Pn in dependency_order:
  IF dependencies_complete:
    1. Load CHUNK-Pn todolist
    
    FOR each TODO in todolist:
      a. Make atomic change
      b. Run todo-specific test
      c. If PASS: commit, mark TODO ✅
      d. If FAIL: STOP, investigate, fix
    END TODO LOOP
    
    2. Update chunk documentation
    3. Mark CHUNK-Pn as IMPLEMENTED
    4. Update research CHUNK-Rn to IMPLEMENTED
    
    IF chunks_processed % 3 == 0 OR context > 35%:
      Context reset (save progress, reload plan)
    END IF
    
    5. Proceed to next ready chunk
  END IF
END CHUNK LOOP

Process

  1. Load Plan Document
    • Read .claude/plans/active/[feature]_plan.md
    • Extract chunk manifest and dependency graph
    • Verify plan status is APPROVED
  2. Determine Execution Order Based on chunk dependency graph:
    • Independent chunks first (parallel capable)
    • Dependent chunks in order
    • Final chunks (e.g., test additions)
  3. For Each Chunk (in dependency order):

    a. Check dependencies complete

    b. Execute each todo atomically:

    • Make single change
    • Run specified test
    • If pass: commit with message
    • If fail: STOP, investigate, fix

    c. After all todos complete:

    • Mark CHUNK-Pn as IMPLEMENTED
    • Update CHUNK-Rn in research to IMPLEMENTED
    • Commit chunk documentation updates

    d. Context management:

    • After every 3 chunks: reload plan
    • If >35% utilization: save, compact, continue
  4. Run Full Test Suite After all chunks complete
  5. Documentation Updates (MANDATORY)
    • Check CODE_TO_WORKFLOW_MAP.md for affected workflows
    • Update workflow files with new line numbers
    • Update function signatures if changed
  6. Context Reset (Every 3 Chunks)
    • Update chunk progress in plan
    • Re-read plan document
    • Verify scope alignment
    • Compact if >35% utilization
  7. Finalize
    • Move plan to .claude/plans/completed/
    • Move research to .claude/research/completed/
    • Run /context-eng:validate to verify

Chunk Status Updates

Update Plan Document

| Chunk | Status | Todos Done | Commit | Research Updated |
|-------|--------|------------|--------|------------------|
| P1 | ✅ IMPLEMENTED | 4/4 | abc123 | ✅ R1 |
| P2 | ▶️ IMPLEMENTING | 2/5 | - | - |

Update Research Document

Mark each CHUNK-Rn status:

  • FOUND → COMPLETE → PLANNED → IMPLEMENTED

Error Handling

Error TypeResponse
Syntax ErrorSTOP. Fix immediately in same todo.
Import ErrorCheck file paths, verify imports.
Test FailureDo NOT add more code. Investigate first.
3+ Failures in chunkMark chunk BLOCKED, try next independent chunk.
3+ Chunks blockedSTOP. Start new session.

Commit Format

Per-todo:

feat(chunk-Pn): Todo N - description
Implements: [feature] chunk N

Per-chunk completion:

feat(chunk-Pn): Complete chunk - [domain]
Completes: CHUNK-Pn, Updates: CHUNK-Rn

Context Budget

  • Plan: 15k tokens
  • Active code (per chunk): ~10k tokens
  • Test results (per chunk): ~5k tokens
  • Max active (3 chunks): ~45k tokens (22.5%)

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

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.