nybble-pr — how to use nybble-pr how to use nybble-pr, nybble-pr setup guide, github cli pull request management, nybble-pr alternative, nybble-pr vs git flow, what is nybble-pr, nybble-pr install, automating github pull requests

v1.0.0
GitHub

About this Skill

Ideal for GitHub-integrated AI Agents seeking automated pull request management and spec artifact analysis. nybble-pr is a skill that automates the generation and management of GitHub pull requests using the GitHub CLI, streamlining developer workflows.

Features

Detects existing pull requests using `gh pr list`
Switches to UPDATE mode if a PR exists, otherwise creates a new one
Uses `git branch --show-current` to get the current branch
Manages pull requests via the GitHub CLI (`gh`)
Generates well-structured pull requests from spec artifacts

# Core Topics

nybblegroup nybblegroup
[0]
[0]
Updated: 3/6/2026

Quality Score

Top 5%
36
Excellent
Based on code quality & docs
Installation
SYS Universal Install (Auto-Detect)
> npx killer-skills add nybblegroup/nybble-agent/nybble-pr
Supports 18+ Platforms
Cursor
Windsurf
VS Code
Trae
Claude
OpenClaw
+12 more

Agent Capability Analysis

The nybble-pr MCP Server by nybblegroup is an open-source Community integration for Claude and other AI agents, enabling seamless task automation and capability expansion. Optimized for how to use nybble-pr, nybble-pr setup guide, github cli pull request management.

Ideal Agent Persona

Ideal for GitHub-integrated AI Agents seeking automated pull request management and spec artifact analysis.

Core Value

Empowers agents to generate well-structured pull requests from spec artifacts and manage them via the GitHub CLI, streamlining development workflows with `gh` commands and JSON data handling.

Capabilities Granted for nybble-pr MCP Server

Automating PR creation from spec artifacts
Updating existing PRs with latest changes
Managing PRs via GitHub CLI with `git branch` and `gh pr list` commands

! Prerequisites & Limits

  • Requires GitHub CLI installation
  • Limited to managing PRs on GitHub
Project
SKILL.md
3.5 KB
.cursorrules
1.2 KB
package.json
240 B
Ready
UTF-8

# Tags

[No tags]
SKILL.md
Readonly
<!-- ⚠️ AUTO-GENERATED by nybble-agent — do not edit directly --> <!-- Source: .nybble/ — run `nybble doctor --fix` to regenerate -->

nybble-pr

Generate a well-structured pull request from the current spec's artifacts and manage it via the GitHub CLI (gh).


1. Detect existing PR

Steps:

  1. Get the current branch: git branch --show-current.
  2. Check for an existing PR: gh pr list --head <branch> --json number,title.
  3. If a PR exists, switch to UPDATE mode. Otherwise, CREATE mode.

2. Gather PR content

Read these files from the active spec directory:

  • spec.md — for the summary and scope.
  • progress.md — for completed tasks.
  • evidence/ — for build, test, coverage, and security scan results.
  • suggestions.md — for any implementation notes.
  • design-context.md — for design references if present.

3. Generate PR description

Structure the description using this format:

markdown
1## Summary 2<!-- 2-3 sentences from spec.md summary --> 3 4## Backend Changes 5<!-- List backend changes from progress.md completed tasks --> 6 7## Frontend Changes 8<!-- List frontend changes from progress.md completed tasks --> 9 10## Automated Testing 11<!-- ONLY reference what the test suite actually validated --> 12<!-- Pull from evidence/test-results.txt and evidence/coverage-summary.txt --> 13- Tests: <passed>/<total> passing 14- Coverage: <N>% lines 15 16## Security 17<!-- Pull from evidence/security-scan.txt if available --> 18 19## Manual Verification 20<!-- Leave as placeholder for human to fill in --> 21- [ ] (Add manual test steps here) 22 23## Spec Reference 24- Spec: `.nybble/specs/active/<feature>/spec.md`

4. Create or update PR

CREATE mode

  • Run: gh pr create --title "<feature title>" --body "<description>"
  • Use the spec's feature name as the PR title (cleaned up for readability).
  • Log a pr_created event to events.jsonl.

UPDATE mode

  • Run: gh pr edit <number> --body "<description>"
  • Preserve any human-added content in the Manual Verification section.
  • Log the update in event details.

5. Post-creation

After the PR is created or updated:

  • Print the PR URL for the human.
  • If evidence shows failing tests or security issues, add a comment noting the known issues.

Trust Behavior

  • L1-L2 (observer, collaborator): Show the full PR description to the human and wait for approval before running gh pr create.
  • L3+ (architect, autonomous): Create the PR directly, then present the URL and description for review.

File Locations

  • Spec artifacts: .nybble/specs/active/<feature>/
  • Event log: .nybble/events.jsonl
  • PR tool: gh CLI (must be installed and authenticated)

What This Skill Does NOT Do

  • Merge PRs — merging is a human decision.
  • Fill in manual verification — that section is for the human.
  • Run CI — it references CI results from evidence, but does not trigger CI pipelines directly.
  • Handle merge conflicts — if conflicts exist, notify the human.

Project Context

  • CORE.md: .nybble/memory/CORE.md
  • Domain files: .nybble/foundation/domains.yaml
  • Design principles: Clarity, Consistency
  • Workflow skills: nybble-spec-create, nybble-spec-edit, nybble-run, nybble-fix, nybble-skill-extract, nybble-conventions, nybble-domains, nybble-design, nybble-security, nybble-events, nybble-verify, nybble-curate-feedback, nybble-pr, nybble-build-verify
  • Project pattern skills: .nybble/skills/ (populated after features ship)

Related Skills

Looking for an alternative to nybble-pr or building a 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

testing

Logo of lobehub
lobehub

Testing is a process for verifying AI agent functionality using commands like bunx vitest run and optimizing workflows with targeted test runs.

73.3k
0
Communication

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