Claude Code Plugins

Community-maintained marketplace

Feedback

Collaborative problem-solving protocols: write technical specifications (spec, or alspec), create implementation plans (plan, or alplan), or use Align-and-Do Protocol (AAD). Also generates PR/MR descriptions (aldescription).

Install Skill

1Download skill
2Enable skills in Claude

Open claude.ai/settings/capabilities and find the "Skills" section

3Upload to Claude

Click "Upload skill" and select the downloaded ZIP file

Note: Please verify skill by going through its instructions before using it.

SKILL.md


name: alignfirst description: Collaborative problem-solving protocols: write technical specifications (spec, or alspec), create implementation plans (plan, or alplan), or use Align-and-Do Protocol (AAD). Also generates PR/MR descriptions (aldescription). license: CC0 1.0 metadata: author: Paleo version: "2.0.2" repository: https://github.com/paleo/alignfirst

AlignFirst Guide

Protocols

Choose the appropriate protocol based on the task:

  • Technical Specification (spec, or alspec): Read spec-protocol.md to write a technical specification
  • Implementation Plans (plan, or alplan): Read plan-protocol.md to create implementation plans from a spec
  • Align-and-Do Protocol (AAD): Read do-protocol.md for smaller tasks without formal spec/plans
  • Description (aldescription): Read description-protocol.md to write a description summarizing implemented work

TASK_DIR Location

TASK_DIR is the directory where work files related to a task are stored. Usually, we use TASK_DIR = _plans/{TICKET_ID}/ (a sub-directory of the _plans folder). If no ticket ID is known, ask the user for it.

  • Create TASK_DIR if it doesn't exist
  • Or, list existing files

File Naming Convention

Format: {CYCLE_LETTER}{FILE_NUMBER}-{FILE_TYPE}.md

Common file types:

  • spec - technical specification
  • plan - implementation plan
  • aad.summary - AAD summary document

Example structure:

_plans/
├── 123/
│   ├── A1-spec.md
│   ├── A2-plan.md
│   └── A3-aad.summary.md
│   └── B1-spec.md

Notes

  • TICKET_ID is a unique identifier for the task, often an issue or ticket number.
  • Cycles are identified by a CYCLE_LETTER (A, B, C...). The user decides when to start a new one.
  • In a cycle, determine the next FILE_NUMBER from existing file names. Every new file must have a bumped file number.
  • Do not bother the user with CYCLE_LETTER or FILE_NUMBER. They are for internal organization. It's up to you to list the files and determine the last CYCLE_LETTER and FILE_NUMBER. Start CYCLE_LETTER with A if there is no existing cycle, and FILE_NUMBER with 1. So you just need to ask for a ticket ID if you don't have one.
  • When the user requests a new cycle: bump CYCLE_LETTER and reset FILE_NUMBER.
  • There is no strict sequence of file types in the workflow. Available file types are also flexible; if you need a new one, just create it.