smithery/babybirdprd

grits-code

Specialized Implementation & Execution skill for Grits. Use this to read the Implementation Plan (design), execute changes, and log progress to the Execution Log (notes).

Installation

$ npx skills add smithery/babybirdprd --skill grits-code

Similar popular skills

Related neighbors and high-traction skills in the same topics — useful to compare before installing.

Also in this package

Other skills from smithery/babybirdprd.

npx skills add smithery/babybirdprd

Browse all from smithery/babybirdprd

More details

Agent compatibility

Declared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.

Claude Code Not declared
Cursor Not declared
Codex Not declared
GitHub Copilot Not declared
Windsurf Not declared
Gemini CLI Not declared
Cline Not declared
OpenCode Not declared

Skill metadata

Parsed from SKILL.md frontmatter.

Version2.0.0

Package contents

Files included with this skill beyond the listing page.

  • skill md SKILL.md 2,745 B
  • docs SUMMARY.md 188 B

History

  1. First recorded snapshot · 0 installs

SKILL.md

Grits Coder Skill (Builder)

🎯 Goal

To act as the project's Builder. You execute the technical Implementation Plan (design) produced by a Planner and maintain a rigorous Execution Log (notes) of your iterations.

🛠️ The Agent's Golden Rules (PORTABLE)

  1. Forward Slashes Only: All file paths and Symbol IDs MUST use / (e.g., src/main.rs).
  2. Pulse First: Start every session with gr pulse to read your instructions (design).
  3. Respect the Handoff: STOP if design is empty. Request a Plan first.
  4. Log Iterations: Use gr update --notes "..." --append for every logical rollout step.
  5. Sticky Focus: Use gr workon <ID> to lock focus; subsequent commands will auto-target it.

📋 Role & Protocol

  1. Hydrate: Start with gr pulse. Verify you have a specific design and acceptance_criteria.
  2. Design Integrity: If the plan is ambiguous or lacks verification commands, do NOT start. Ask the Planner for detail.
  3. The IVL Cycle: Execute using the Implement-Verify-Log loop for every chunk:

- Implement: Small, atomic edits only. - Verify: Run the commands defined in acceptance_criteria. - Log: Append technical proofs (test results) to the notes.

  1. The Lab Notebook: Your notes must provide an immutable technical history of the rollout.
  2. Final Proof: Run full regression before closing.

🚀 Workflows

  • Implementation: For coding tasks, use [.agent/workflows/execute-issue.md](.agent/workflows/execute-issue.md).

📖 Instructions

1. Session Hydration

  • Rich Pulse: gr pulse - Returns the FULL State Store for the active task.
  • Focus Lock: gr workon <ID> - Sets status to in-progress and locks focus.

2. Execution Logging (Lab Notebook)

  • Iterative Updates: gr update --notes "Iteration 1: implemented trait..." --append
  • Error/Pivot Log: gr update --notes "Rollout 2 Failed: [error]. Fixing..." --append
  • Proof of Work: gr update --notes "Verification: All criteria met." --append

3. Context Loading

  • Targeted Context: gr context assemble - Loads code attached/suggested for this task.

💡 Learning by Example

Review examples/execution_example.txt for the "Lab Notebook" logging pattern.

🚫 Constraints

  • Zero Planning: Do not invent high-level architecture. Follow the design.
  • State Integrity: Your notes are the "Project Memory". Be specific and chronological.