skills.rest
redblacktree
redblacktreeCOMMUNITY
Explore all skills by this author →
GitHub

ff_create_plan

Creates actionable implementation plans from Fastflow goals, research, and Codex subagent findings.

1|1|Updated Feb 3, 2026
Search Tags:#planning#markdown#codex#subagents#multi-agent#implementation plan#fastflow
One-click install
npx skills add https://github.com/redblacktree/fastflow --skill ff-create-plan-redblacktree
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ff_create_plan
Source: https://github.com/redblacktree/fastflow/tree/main/internal/templates/files/.codex/skills/ff_create_plan
Command: npx skills add https://github.com/redblacktree/fastflow --skill ff-create-plan-redblacktree

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Turning a development goal into a concrete, executable implementation plan requires reading research, inspecting code, and coordinating findings—work that is error-prone when done ad hoc. This Skill standardizes that process so a fresh agent can implement the plan without hidden context. ## Core Features & Use Cases - Structured Plan Generation: Produces markdown plans under thoughts/shared/plans/ with goal, current state, steps, verification commands, and risks. - Subagent-Orchestrated Research: Delegates bounded, read-only discovery tasks to Codex subagents while keeping architecture decisions local. - Use Case: Given a Fastflow goal like "Add user authentication" with prior research notes, generate a step-by-step plan listing exact files to change, verification commands, and PR notes for the implementation stage. ## Quick Start Ask the agent to create an implementation plan for the current Fastflow goal using any available research in thoughts/shared/research/.

Frequently Asked Questions about ff_create_plan

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create an implementation plan from a Fastflow goal?▼

Read the goal file and any matching research in thoughts/shared/research/, inspect the relevant code, then write a structured markdown plan under thoughts/shared/plans/ following the defined template with steps, files, and verification commands.

How should Codex subagents be used during planning?▼

Use subagents for specific, read-only discovery tasks such as finding existing patterns, analyzing impact, or identifying test commands. Keep final architecture decisions and plan synthesis local, and wait for subagent results before finalizing.

What format does the implementation plan file use?▼

Plans are markdown files named YYYY-MM-DD-{ticket-or-topic}.md containing sections for goal, current state, desired end state, out of scope, subagent findings, numbered implementation steps with file paths, verification commands, branch/PR notes, and risks.

When should I ask for clarification instead of planning?▼

Ask for clarification only when the repository and available skills cannot answer a decision that blocks implementation. The final plan must not contain open questions, so resolve uncertainties through code inspection or subagent research first.

What are the limitations of delegating planning to subagents?▼

Subagents should not receive the whole planning task; they handle bounded, read-only research slices only. Delegating the full plan loses synthesis quality, and important facts returned by subagents must be verified before inclusion.

Related Skills

Explore More Software Engineering Skills→
unblockedunblocked
13

headless-enrich-ticket

Gather context and attach structured enrichment documents to issue tracker tickets.

Official
Advanced
openaiopenai
5.3k

executing-plans

Executes written implementation plans with review checkpoints and verification steps.

Official
Intermediate
suzbotsuzbot
5

implement-feature

Implement planned software features using TDD with human testing checkpoints.

Community
Advanced
ThiagoVitaleThiagoVitale

gsd-fast

Execute trivial tasks directly in the current context without subagents.

Community
Basic