~ / rigs / getCurrentThread / orchestration-kit

getCurrentThread/orchestration-kit

Claude Code 플러그인: Orca CLI 기반의 디스패치 게이트와 장부, 마스터 승계, 리뷰 워크플로 (Claude Code plugin: dispatch gate and ledger, master succession and a review workflow on top of the Orca CLI)

↗ GitHub ★ 0 MIT updated 1d ago project Claude CodeCodexCursor Claude plugin
field: unknown workflow: PR / code reviewDocs
stack: Python
share on X
ARCHETYPE
Automator
Hooks on every lifecycle event: format, lint, notify, log.
CONTEXT TAX · EVERY TURN
~7.0k tokens
Heavy · median rig: 2.3k · breakdown
GUARDRAILS
2/5
Pre-tool screening hook · No YOLO mode · details
Quality 29/100 Guardrails v2

Copy this rig

# review before running: this installs third-party code
$ claude mcp add code-review-graph -e PYTHONUTF8=YOUR_PYTHONUTF8 -- code-review-graph serve
$ npx degit getCurrentThread/orchestration-kit/harness/.claude ./rig-orchestration-kit  # inspect, then merge into .claude/

MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.

$ claude mcp add code-review-graph -e PYTHONUTF8=YOUR_PYTHONUTF8 -- code-review-graph serve
$ npx degit getCurrentThread/orchestration-kit/harness/.agents/skills/daily-triage .claude/skills/daily-triage
$ npx degit getCurrentThread/orchestration-kit/harness/.agents/skills/postmortem .claude/skills/postmortem
$ npx degit getCurrentThread/orchestration-kit/harness/.agents/skills/scope .claude/skills/scope
$ npx degit getCurrentThread/orchestration-kit/skills/master .claude/skills/master
$ curl -fsSL --create-dirs -o .claude/agents/code-reviewer.md https://raw.githubusercontent.com/getCurrentThread/orchestration-kit/master/agents/code-reviewer.md
$ curl -fsSL --create-dirs -o .claude/agents/silent-failure-hunter.md https://raw.githubusercontent.com/getCurrentThread/orchestration-kit/master/agents/silent-failure-hunter.md

Merge into .claude/settings.json (project) or ~/.claude/settings.json (user). Hook commands reference scripts in the source repo — copy those too.

{
  "hooks": {
    "PreToolUse": [
      {
        "matcher": "Write|Edit",
        "hooks": [
          {
            "type": "command",
            "command": "bash \"$CLAUDE_PROJECT_DIR/.claude/hooks/no-emoji-check.sh\""
          }
        ]
      }
    ]
  }
}

MCP servers (1)

serversourceest. tokens
code-review-graph
env: PYTHONUTF8
binary 2.5k

Skills (4)

Subagents (2)

code-reviewer
model: sonnet
Code review specialist for quality, security and maintainability. Used by the orchestration-review workflow; invoke it directly only when a code review is requested.
silent-failure-hunter
model: sonnet
Review code for silent failures, swallowed errors, bad fallbacks, and missing error propagation.

Hooks (8)

eventmatcherruns
PreToolUseWrite|Editbash "$CLAUDE_PROJECT_DIR/.claude/hooks/no-emoji-check.sh"
PostToolUseEdit|Write|Bashcommand -v code-review-graph >/dev/null 2>&1 || exit 0; PYTHONUTF8=1 code-review-graph update --skip-flows
SessionStart*command -v code-review-graph >/dev/null 2>&1 || exit 0; code-review-graph status 2>&1 || echo 'code-review-graph: no status. If the line above says that no graph was found, the first Edit, Write or Bash call builds it, or build it now with:
SessionStart*bash "$CLAUDE_PROJECT_DIR/.claude/hooks/setup-check.sh"
SessionStart*bash "$CLAUDE_PROJECT_DIR/.claude/hooks/memory-health.sh"
SessionStart*bash "$CLAUDE_PROJECT_DIR/.claude/hooks/load-context.sh"
SessionStart*bash "$CLAUDE_PROJECT_DIR/.claude/hooks/wiki-health.sh"
UserPromptSubmit*bash "$CLAUDE_PROJECT_DIR/.claude/hooks/memory-selfcheck.sh"

Slash commands (2)

/scope/orchestration-review

Cursor rules (1)

agentic-router · always

Similar rigs

copied ✓