grp06/useful-codex-skills/implement-execplan
Claude Code and agent skill by grp06/useful-codex-skills
- community
Checking that the SKILL.md file is still in the repository.
Execute an ExecPlan from a work item under `.agent/work/` or, when necessary, from the older singleton `.agent/execplan-pending.md` flow. Use when the user asks to implement a plan, execute the pending plan, or resume blocked implementation work.
Install
npx vskill@latest install grp06/useful-codex-skills/implement-execplanSKILL.md
Implement ExecPlan
Use the ExecPlan as the implementation contract for delivering both the intended behavior and the intended simplification of the system.
Preferred Input Resolution
Preferred target resolution order:
- explicit work-item path supplied by the user
- explicit
execplan.mdpath supplied by the user .agent/activewhen it points to a work item with:
stage="plan"andstate="completed", orstage="implementation"andstate="blocked"
- the most recently updated work item under
.agent/work/matching those same rules - legacy fallback:
.agent/execplan-pending.md
If no supported plan exists, stop and tell the user.
Work Item Responsibilities
If operating on a work item, read:
meta.jsondecision.mdwhen presentexecplan.md
This skill owns execution-state transitions in meta.json:
- when starting work:
stage="implementation",state="active" - when blocked or only partially complete:
stage="implementation",state="blocked" - when implementation completes:
stage="implementation",state="completed"
Do not rename plans or move directories to represent lifecycle state.
Ousterhout Lens
When the plan leaves room for judgment:
- prefer deep modules over shallow wrappers
- prefer interfaces that hide sequencing and policy
- prefer fewer concepts, fewer knobs, and fewer special cases
- prefer simple mental models over clever decomposition
Workflow
- Read the ExecPlan in full, then read
.agent/PLANS.md. - Reconstruct the intended behavior, target boundary, and complexity dividend.
- If available, read
decision.mdso the implementation stays aligned with the original decision rationale. - Before coding, update
meta.jsontostage="implementation"andstate="active". - Inspect the relevant code paths before editing.
- Implement milestone by milestone.
- Keep the ExecPlan's
Progress,Surprises & Discoveries,Decision Log, andOutcomes & Retrospectivesections up to date as you go. - After each meaningful slice, run the plan's validation steps or the nearest targeted verification.
- If you finish the implementation, set
state="completed". - If you cannot finish safely in the current turn, record the blocker in the plan, set
state="blocked", and stop cleanly.
Implementation-First Rule
This skill is for execution, not more planning.
You may update the ExecPlan during implementation only to:
- record progress actually made
- record discoveries from code already inspected
- tighten milestones or acceptance criteria to match current repo reality
- document design decisions that unblock the next implementation step
Do not turn an implementation turn into another plan-improvement pass.
Anti-Patterns
- satisfying the letter of the plan while preserving the same interface burden
- adding wrappers, adapters, or helper layers that hide little
- pushing sequencing or policy outward to callers
- finishing with only ExecPlan edits when implementation work remains
- using file moves or renames as the primary state transition
Security
- Security scanPassed, 100 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingimplement-execplan/SKILL.md
Popularity
- Repository stars
- 34
- Repository forks
- 4
- 7-day trend score
- 18
Stars and forks are for the whole GitHub repository, not this skill alone. Refreshed 18 Mar 2026, 7 months ago.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/grp06/useful-codex-skills/implement-execplan)