create-implementation-plan
Explicitly invoked workflow for creating a concrete implementation plan file from a requested task.
Skill by KFlamsted
Chalk Labs
Need help deploying create-implementation-plan?
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/kflamsted-create-implementation-plan
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: 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. Checked Oct 25, 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/create-implementation-plan
What's inside
- Skill name
create-implementation-plan- Description
- Explicitly invoked workflow for creating a concrete implementation plan file from a requested task.
- License
- MIT
- Instructions
- 36 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Create Implementation Plan Create a detailed implementation plan only. Do not change code. ## Input - A task description from the user. ## Process 1. Inspect the repository enough (through code exploration, tool calls, documentation and/or existing plans) to understand the task. 2. Before planning, perform a requirements and ambiguity audit. Consider: - the desired outcome and how success will be evaluated; - scope, non-goals, and boundaries; - user-visible behavior and workflows; - edge cases, failure behavior, and validation; - compatibility, migration, API, and data implications; - operational, security, performance, and rollout constraints; - testing and acceptance criteria. 3. Identify every unresolved decision that could change the plan, implementation, public behavior, or acceptance criteria. Do not infer the user's preference merely because one option seems conventional. Repository evidence may answer factual questions, but it must not substitute for product or scope decisions. Avoid questions already answered by repository conventions or concerning inconsequential implementation details. 4. Ask focused clarification questions before writing the plan. Explain briefly why each question matters and provide concrete options with a recommended default when practical. Use open-ended questions when predefined options would be limiting or misleading. 5. After each response, update your understanding and look for newly exposed ambiguities. Continue asking follow-up questions until: - the intended outcome and boundaries are explicit; - materially different interpretations have been resolved; - acceptance criteria are testable; and - the resulting plan can be implemented without product or scope decisions being made by the implementer. 6. Summarize the agreed requirements, decisions, non-goals, and remaining assumptions. Ask the user to confirm or correct this summary. Do not create the plan until it is confirmed. 7. Create a concrete implementation plan that is detailed enough so a low- to mid level engineer, without any prior repository context, can implement consistently with high quality. For questions and clarifications, use the available clarification tool (`AskQuestion`, `question`, or similar variants such as `AskUserQuestions`) if available; otherwise, ask in plain text. ## Output - Write the plan to `<TASK_NAME>_PLAN.md` unless project rules determine something else. Then go by project rules. - Make sure the chosen plan name does not overwrite an existing plan. - Include relevant files, implementation steps, tests, risks, and assumptions. - Make sure the plan is detailed enough so a low- to mid level engineer, without any prior repository context, can implement it consistently, with high quality, without further questions.
From KFlamsted/agent-workflow-kit, MIT.
More from KFlamsted/agent-workflow-kit
- 0.561score
implement-plan-loop
KFlamsted
Starts a delegated implementation loop from an implementation plan or freeform task prompt.
SkillAFreshness: AClaude Code +9 - 0.529score
hology-game-engine
KFlamsted
Reference for building games with the Hology engine (a TypeScript/Three.js game framework).
SkillBFreshness: BClaude Code +9 - 0.521score
end-to-end-orchestrator
KFlamsted
Orchestrates software tasks end to end by delegating planning, phased implementation, review, and final reporting to specialized subagents.
SkillBFreshness: BClaude Code +9 - 0.521score
improve-human-spec-review
KFlamsted
Guide a human through reviewing a specification or implementation plan in clear, meaningful chunks without reviewing or editing the document for them.
SkillBFreshness: BClaude Code +9 - 0.521score
review-findings-to-plan
KFlamsted
Explicitly invoked workflow for discussing review findings one by one and converting agreed fixes into an implementation plan.
SkillBFreshness: BClaude Code +9 - 0.521score
code-review
KFlamsted
Explicitly invoked workflow for thoroughly reviewing the current branch against master or main without making code changes.
SkillBFreshness: BClaude Code +9 - 0.521score
clean-solid-review-loop
KFlamsted
Runs a Clean/SOLID quality review/fix loop until a 10/10 score, with a required behavior code-reviewer after each implementer.
SkillBFreshness: BClaude Code +9 - 0.521score
end-to-end-orchestrator-detailed
KFlamsted
Orchestrates a software task through Gherkin behaviour planning, research, optional architecture, an INVEST implementation plan, then implement/review with…
SkillBFreshness: BClaude Code +9 - 0.518score
code-implementation-loop-orchestrator
KFlamsted
Coordinates an implementation/review loop by delegating all code writing and reviewing to subagents.
SkillBFreshness: BClaude Code +9 - 0.518score
code-review-loop
KFlamsted
Runs a reviewer-first code review/fix loop by delegating review and implementation to dedicated subagents.
SkillBFreshness: BClaude Code +9 - 0.518score
commit-message-generate
KFlamsted
Generates a concise commit message from provided changes without running git commands.
SkillBFreshness: BClaude Code +9 - 0.513score
implement-plan
KFlamsted
Explicitly invoked workflow for implementing an existing plan file in the repository.
SkillBFreshness: BClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Sep 12, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/KFlamsted/agent-workflow-kit
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from KFlamsted
- 0.561score
implement-plan-loop
KFlamsted
Starts a delegated implementation loop from an implementation plan or freeform task prompt.
SkillAFreshness: AClaude Code +9 - 0.529score
hology-game-engine
KFlamsted
Reference for building games with the Hology engine (a TypeScript/Three.js game framework).
SkillBFreshness: BClaude Code +9 - 0.521score
review-findings-to-plan
KFlamsted
Explicitly invoked workflow for discussing review findings one by one and converting agreed fixes into an implementation plan.
SkillBFreshness: BClaude Code +9 - 0.521score
code-review
KFlamsted
Explicitly invoked workflow for thoroughly reviewing the current branch against master or main without making code changes.
SkillBFreshness: BClaude Code +9 - 0.521score
end-to-end-orchestrator
KFlamsted
Orchestrates software tasks end to end by delegating planning, phased implementation, review, and final reporting to specialized subagents.
SkillBFreshness: BClaude Code +9 - 0.521score
clean-solid-review-loop
KFlamsted
Runs a Clean/SOLID quality review/fix loop until a 10/10 score, with a required behavior code-reviewer after each implementer.
SkillBFreshness: BClaude 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 create-implementation-plan 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
create-implementation-plan skill. Agent Mart. https://agentmart.dev/skills/kflamsted-create-implementation-plan (last updated 2026-10-24)