KS
Killer-Skills

implementation-planning — how to use implementation-planning how to use implementation-planning, implementation-planning setup guide, implementation-planning vs competing skills, what is implementation-planning, implementation-planning alternative for AtlasMemory, implementation-planning install for AI agents, implementation-planning tutorial for developers, implementation-planning best practices for build execution

v1.0.0
GitHub

About this Skill

Ideal for AI Agents like Cursor, Windsurf, or Claude Code that require detailed build execution planning with file deltas and workstreams. Implementation-planning is a skill that creates the Implementation Plan section for build execution, returning a draft section to the orchestrator.

Features

Generates exhaustive file deltas with change type and explicit owner
Creates workstreams with dependencies, merge points, and explicitly owned files
Produces an integration/merge points checklist for smooth execution
Runs as a sub-agent to return a draft section to the orchestrator
Enforces implementation plan best practices for AtlasMemory development

# Core Topics

Atlas-Memory-Framework Atlas-Memory-Framework
[0]
[0]
Updated: 3/7/2026

Quality Score

Top 5%
54
Excellent
Based on code quality & docs
Installation
SYS Universal Install (Auto-Detect)
Cursor IDE Windsurf IDE VS Code IDE
> npx killer-skills add Atlas-Memory-Framework/AtlasMemory-Cursor-Tools/implementation-planning

Agent Capability Analysis

The implementation-planning MCP Server by Atlas-Memory-Framework is an open-source Categories.community integration for Claude and other AI agents, enabling seamless task automation and capability expansion. Optimized for how to use implementation-planning, implementation-planning setup guide, implementation-planning vs competing skills.

Ideal Agent Persona

Ideal for AI Agents like Cursor, Windsurf, or Claude Code that require detailed build execution planning with file deltas and workstreams.

Core Value

Empowers agents to generate exhaustive file deltas with change type, explicit owner, and rationale, while also creating workstreams with dependencies and merge points, utilizing AtlasMemory for efficient development workflows.

Capabilities Granted for implementation-planning MCP Server

Automating implementation plan generation for build execution
Generating exhaustive file deltas with explicit ownership and rationale
Creating workstreams with dependencies, merge points, and owned files for efficient development

! Prerequisites & Limits

  • Requires AtlasMemory for development workflows
  • Must be run as a sub-agent to return draft implementation plans to the orchestrator
Project
SKILL.md
4.2 KB
.cursorrules
1.2 KB
package.json
240 B
Ready
UTF-8

# Tags

[No tags]
SKILL.md
Readonly

/implementation-planning

Purpose

Create the Implementation Plan section for build execution. Run as a sub-agent and return a draft section to the orchestrator; do not write the plan artifact directly.

Required outputs

  • Exhaustive file deltas with change type, explicit owner (existing agent/WS), and rationale.
  • Workstreams with dependencies, merge points, and explicitly owned files.
  • An integration / merge points checklist (what gets integrated, how, and what gates run).
  • Enforce the workstream file-ownership rule: each file delta is owned by exactly one workstream until an explicit merge point.
  • Phases and tasks mapped to workstreams/owners.
  • Evidence-based exit criteria per phase.
  • Build-time gates for each phase.
  • Test Plan including at least a minimal test matrix (risk -> test type -> where it runs).
  • Rollout/Deployment steps (even minimal) and an explicit rollback trigger + rollback steps.
  • Draft section content for ## Implementation Plan

Anti-placeholder rule (hard rule)

  • Do not use placeholder language like “run smoke tests” or “add gates” without naming:
    • the gate name(s),
    • where they run (CI vs local vs deployed), and
    • the entrypoint/command (or test runner/target) and what “green” means.
  • If specifics are truly unknown, convert them into either:
    • an explicit Decision boundary (A/B/C) with a recommended default, OR
    • a DR-backed Defer with an explicit trigger.

Q/A wrapper

  • The orchestrator must run the inline Q/A loop to confirm user understanding and agreement on scope, ownership, and gates.

Ownership policy

  • Prefer existing agents from the plan's ## Context Snapshot or the user-provided roster.
  • If no roster is available, return Questions asking for the available agents/owners before assigning work.
  • Do not default all ownership to the user unless the user explicitly requests it.

Parallel workstreams / multi-agent contract (hard rule)

  • If the user requests multiple agents or parallel workstreams:
    • Ensure the draft includes an explicit agent roster (names/handles) and assigns ownership for each workstream and file-delta cluster.
    • If the roster is missing, ask a single focused question to obtain it (do not guess).
    • The orchestrator should mirror this roster into ## Context Snapshot (call this out explicitly in Notes if needed).

Sub-agent output contract

Return a single block in this shape:

md
1DraftSection: 2<exact section content for ## Implementation Plan (must include the section header)> 3 4Checklist: 5- <criterion>: Pass | Fail 6 7Questions: 8- <if blocked> 9 10Notes: 11- <optional risks/assumptions/tests updates>

Malformed output handling

  • If you cannot produce the exact section header or required fields, return Questions explaining what is missing and leave DraftSection as N/A.

User experience rule (no "go read the plan")

  • When asking the user to confirm plan readiness, paste the key parts directly in the chat response:
    • File deltas (owned)
    • Workstreams + owned files + merge points
    • Phase list with exit criteria
    • Test matrix + rollback

Output template (required for PlanTier: Full)

Use this exact structure so the /plan validator can deterministically check executability:

md
1## Implementation Plan 2### Agent roster (required for PlanTier: Full) 3- <agent/owner>: <responsibilities> 4 5### File Deltas (exhaustive) + rationale 6- path/to/file.ext - change type (create/modify/delete) - owner (WSx / agent) - rationale 7 8### Workstreams + merge points 9- WS1: <name> 10 - Owner: 11 - Depends on: 12 - Review gates (named): 13 - G-... 14 - Owns files: 15 - path/to/file.ext 16 - Merge point / integration step: 17 18### Phases + tasks + exit criteria 19#### Phase 1: <name> 20- Owner(s): 21- Depends on: 22- Tasks (by owner): 23 - Owner: <agent> 24 - [ ] Task 25- Exit criteria (evidence): 26- Gates (named): 27 - G-... 28 29### Review gates (named + definitions) 30- G-...: 31 - Where it runs: CI | Local | Deployed 32 - Entry point / command: 33 - Green means: 34 35### Merge points -> required gates 36- MP1: <merge point> 37 - Blocks on: 38 - G-... 39 40### Test Matrix 41- Area/component - risk - test type - where it runs 42 43### Test plan (CI vs deployed) 44- CI: 45 - ... 46- Deployed environment: 47 - ... 48 49### Rollout / Rollback 50- Rollout: 51- Rollback trigger: 52- Rollback steps:

Related Skills

Looking for an alternative to implementation-planning or building a Categories.community AI Agent? Explore these related open-source MCP Servers.

View All

widget-generator

Logo of f
f

widget-generator is an open-source AI agent skill for creating widget plugins that are injected into prompt feeds on prompts.chat. It supports two rendering modes: standard prompt widgets using default PromptCard styling and custom render widgets built as full React components.

149.6k
0
Design

chat-sdk

Logo of lobehub
lobehub

chat-sdk is a unified TypeScript SDK for building chat bots across multiple platforms, providing a single interface for deploying bot logic.

73.0k
0
Communication

zustand

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

72.8k
0
Communication

data-fetching

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

72.8k
0
Communication