agentrigs
+ submit rig
~ / rigs / Dana-Afazeli / agentic-swe-kit

Dana-Afazeli/agentic-swe-kit

A maintained template for the agentic-SWE harness: proven gates, Claude Code hooks, CI, a headless review loop, and the workflow docs behind them. Use this template, run kit.py init, take updates with kit.py update.

↗ GitHub ★ 0 MIT updated 1d ago project Claude CodeCodex
share on X
ARCHETYPE
Fort Knox
Deny lists, pre-tool hooks, sandboxing. Nothing touches prod without a signature.
CONTEXT TAX · EVERY TURN
~1.7k tokens
Featherweight · median rig: 2.3k · breakdown
GUARDRAILS
4/5
Blocks destructive commands · Pre-tool screening hook · No YOLO mode · Sandbox or ask-first rules · details

Copy this rig

# review before running: this installs third-party code
$ npx degit Dana-Afazeli/agentic-swe-kit/.claude ./rig-agentic-swe-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.

$ npx degit Dana-Afazeli/agentic-swe-kit/.claude/skills/review-loop .claude/skills/review-loop
$ curl -fsSL --create-dirs -o .claude/agents/plan-reviewer.md https://raw.githubusercontent.com/Dana-Afazeli/agentic-swe-kit/main/.claude/agents/plan-reviewer.md

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

{
  "permissions": {
    "deny": [
      "Bash(git push --force:*)",
      "Bash(git push -f:*)",
      "Bash(git push origin main)",
      "Bash(git push -u origin main)",
      "Bash(git push origin HEAD:main)",
      "Bash(git reset --hard:*)",
      "Bash(rm -rf ~:*)",
      "Bash(git commit --no-verify:*)",
      "Bash(gh label:*)",
      "Bash(gh pr edit --add-label:*)",
      "Bash(gh pr edit --remove-label:*)"
    ],
    "ask": [
      "Bash(gh pr merge:*)",
      "Edit(/.claude/**)",
      "Edit(/.github/**)",
      "Edit(/scripts/**)",
      "Edit(/kit.py)",
      "Edit(/Makefile)",
      "Edit(/pyproject.toml)",
      "Edit(/uv.lock)",
      "Edit(/.pre-commit-config.yaml)",
      "Edit(/.python-version)",
      "Edit(/.gitignore)",
      "Edit(/.betterleaks.toml)",
      "Edit(/.betterleaksignore)",
      "Edit(/GNUmakefile)",
      "Edit(/makefile)",
      "Edit(/pytest.toml)",
      "Edit(/.pytest.toml)",
      "Edit(/pytest.ini)",
      "Edit(/.pytest.ini)",
      "Edit(/ruff.toml)",
      "Edit(/.ruff.toml)",
      "Edit(/pyrightconfig.json)",
      "Edit(/.coveragerc)",
      "Edit(/.coveragerc.toml)",
      "Edit(/.importlinter)",
      "Edit(/setup.cfg)",
      "Edit(/tox.ini)",
      "Edit(/uv.toml)"
    ]
  },
  "hooks": {
    "PreToolUse": [
      {
        "matcher": "Bash",
        "hooks": [
          {
            "type": "command",
            "command": "uv run --project \"$CLAUDE_PROJECT_DIR\" python \"$CLAUDE_PROJECT_DIR/scripts/guard_bash.py\""
          }
        ]
      }
    ]
  }
}

Skills (1)

review-loop

Subagents (1)

plan-reviewer
model: sonnet
Conformance review of a pull request against its brief. Give it the path of the brief and the PR number. It reports only gaps between the two — an acceptance criterion with no test that bites, a chang

Hooks (3)

eventmatcherruns
PreToolUseBashuv run --project "$CLAUDE_PROJECT_DIR" python "$CLAUDE_PROJECT_DIR/scripts/guard_bash.py"
PostToolUseEdit|Writeuv run --project "$CLAUDE_PROJECT_DIR" python "$CLAUDE_PROJECT_DIR/scripts/fmt_hook.py"
Stop*uv run --project "$CLAUDE_PROJECT_DIR" python "$CLAUDE_PROJECT_DIR/scripts/stop_gate.py"

Permissions

deny (11)
Bash(git push --force:*)
Bash(git push -f:*)
Bash(git push origin main)
Bash(git push -u origin main)
Bash(git push origin HEAD:main)
Bash(git reset --hard:*)
Bash(rm -rf ~:*)
Bash(git commit --no-verify:*)
Bash(gh label:*)
Bash(gh pr edit --add-label:*)
Bash(gh pr edit --remove-label:*)
ask (28)
Bash(gh pr merge:*)
Edit(/.claude/**)
Edit(/.github/**)
Edit(/scripts/**)
Edit(/kit.py)
Edit(/Makefile)
Edit(/pyproject.toml)
Edit(/uv.lock)
Edit(/.pre-commit-config.yaml)
Edit(/.python-version)
Edit(/.gitignore)
Edit(/.betterleaks.toml)
Edit(/.betterleaksignore)
Edit(/GNUmakefile)
Edit(/makefile)
Edit(/pytest.toml)
Edit(/.pytest.toml)
Edit(/pytest.ini)
Edit(/.pytest.ini)
Edit(/ruff.toml)
Edit(/.ruff.toml)
Edit(/pyrightconfig.json)
Edit(/.coveragerc)
Edit(/.coveragerc.toml)
Edit(/.importlinter)
Edit(/setup.cfg)
Edit(/tox.ini)
Edit(/uv.toml)
allow (25)
Bash(make:*)
Bash(uv run:*)
Bash(uv sync:*)
Bash(uv add:*)
Bash(uv lock:*)
Bash(uv python:*)
Bash(git status:*)
Bash(git diff:*)
Bash(git log:*)
Bash(git show:*)
Bash(git branch --show-current)
Bash(git fetch:*)
Bash(git pull)
Bash(git add:*)
Bash(git commit:*)
Bash(git switch main-:*)
Bash(git switch -c main-:*)
Bash(git push -u origin main-:*)
Bash(git push origin main-:*)
Bash(gh pr create --base main:*)
Bash(gh pr view:*)
Bash(gh pr checks:*)
Bash(gh pr ready:*)
Bash(gh pr diff:*)
Bash(gh run view:*)

Similar rigs

spxrogers/agentsync
Sync AI coding-agent configs (Claude Code, OpenCode, Codex, and more) from one canonical, committable source.
★ 32 setup Claude CodeCodex 1 skills 2d ago
Pragmatist 12.8k tok ·
HGInsights/claude-code-setup
Production-tested Claude Code harness configuration: hooks, skills, subagents, and automated review loops
★ 2 setup Claude CodeCodexGemini CLI 2 agents 17d ago
Pragmatist 1.9k tok ·
RemiAsselin42/claude-config
Memory, features, skills, and agents for Claude, centralized and versioned, one-command installation
★ 1 setup Claude Code 3 agents 13 hooks 6d ago
Automator 2.1k tok ·
code-yeongyu/my-claude-code-harness
I'm confident this setup will become deprecated and outdated in less than 3 months
★ 12 setup Claude Code 3 skills 3 agents 3 hooks 11mo ago
Pragmatist 2.8k tok ·
copied ✓