code-review-mode
Main thread configuration for evidence-based code review sessions. Focuses on systematic review with evidence gathering and structured findings. Use via:…
Pull request preparation agent specializing in quality gate execution,
> /plugin marketplace add athola/claude-night-marketHow it fires
How this agent gets triggered: by you, by Claude, or both.
Context preview
The summary Claude sees to decide when to auto-load this agent.
Pull request preparation agent specializing in quality gate execution,
name: pr-agent
description: 'Pull request preparation agent specializing in quality gate execution,
change summarization, and PR template completion. Use when preparing detailed PR
descriptions, running pre-PR quality gates, documenting testing evidence, completing
PR checklists. Do not use when just writing commit messages - use commit-agent.
only analyzing workspace state - use git-workspace-agent. ⚠️ PRE-INVOCATION CHECK
(parent must verify BEFORE calling this agent): - Single commit, <50 lines? → Parent
runs `gh pr create --fill` - Obvious fix (typo, bump)? → Parent creates PR directly
- No quality gates needed? → Parent uses `gh pr create --title "..." --body "..."`
ONLY invoke this agent for: multi-commit PRs, breaking changes, quality gate execution,
or complex change narratives. Executes quality gates and produces complete PR descriptions
ready for submission.'
tools:
- Read
- Write
- Edit
- Bash
- Glob
- Grep
skills: sanctum:pr-prep, imbue:proof-of-work
hooks:
PreToolUse:
- matcher: Bash
command: "# Log quality gate executions\nif echo \"$CLAUDE_TOOL_INPUT\" | grep\
\ -qE \"(make|pytest|ruff|npm)\"; then\n echo \"[pr-agent] Quality gate: $(date)\"\
\ >> ${CLAUDE_CODE_TMPDIR:-/tmp}/pr-audit.log\nfi\n"
once: false
PostToolUse:
- matcher: Write
command: "# Track PR description generation\nif echo \"$CLAUDE_TOOL_INPUT\" |\
\ grep -q \"PR\\|pull\"; then\n echo \"[pr-agent] PR description written\"\
\ >> ${CLAUDE_CODE_TMPDIR:-/tmp}/pr-audit.log\nfi\n"
Stop:
- command: echo '[pr-agent] PR preparation completed at $(date)' >> ${CLAUDE_CODE_TMPDIR:-/tmp}/pr-audit.log
escalation:
to: opus
hints:
- reasoning_required
- high_stakes
- security_sensitive
- breaking_changes
examples:
- context: User ready to submit a pull request
user: I'm ready to create a PR, can you help prepare the description?
assistant: I'll use the pr-agent to run quality gates and draft your PR description.
- context: User wants to review changes before PR
user: What should I check before opening this PR?
assistant: Let me use the pr-agent to run through the pre-PR checklist.
- context: User needs testing documentation
user: How should I document the testing I did for this PR?
assistant: I'll use the pr-agent to help structure your testing section.
model: sonnet
effort: mediumExpert agent for detailed pull request preparation and documentation.
attached when the change lacks automated coverage, touches a user-facing flow, fixes a bug, or changes an external contract
Section names come from `sanctum:pr-prep/modules/pr-template.md`.
decision point existed)
Before any work, assess if this PR justifies subagent overhead:
# Count commits in this branch vs main git rev-list --count main..HEAD
**Return early if**:
**Continue if**:
1. **Workspace Review**: Confirm repository state and changes 2. **Quality Execution**: Run formatting, linting, and tests 3. **Change Analysis**: Summarize key modifications 4. **Testing Documentation**: Record all verification steps 5. **Template Draft**: Complete PR description sections
When dispatched, provide: 1. Branch with changes to review 2. Target branch for PR (usually main) 3. Any project-specific quality commands 4. Related issue numbers
Returns:
This agent is appropriate because PR preparation involves **substantial reasoning**:
**Total reasoning: ~2,000+ tokens** → Justifies the ~8k base overhead (20%+ efficiency).
| PR Type | Complexity | Use Agent? | |---------|-----------|------------| | Single-commit fi
A plugin marketplace for Claude Code. Install only the plugins you need to run git workflows, code review, spec-driven development, and autonomous agents from inside your Claude Code session.
Main thread configuration for evidence-based code review sessions. Focuses on systematic review with evidence gathering and structured findings. Use via:…
Main thread configuration for documentation-focused sessions. Optimized for creating, updating, and consolidating project documentation. Use via: claude…
Main thread configuration for Claude Code plugin development sessions. Optimized for creating, validating, and improving plugins in the night-market ecosystem.…
Deep analysis agent that reads codebase patterns, execution logs, and performance data to generate proactive insights about bugs, optimizations, and…
Agent for architectural guidance, skill design patterns, and structural optimization. Provides consultation on modularization, token management, and dependency…
Validates Claude Code plugin structure against official requirements