~ / rigs / aravindhsampath / agentic-template

aravindhsampath/agentic-template

Cookie-cutter project template for an agentic development workflow: Claude Code architect + engineer subagents, Pi-based multi-model code review (Deepseek + Kimi), strict branch hygiene, declarative apply-standard script.

↗ GitHub ★ 2 no license updated 4mo ago project Claude Code
share on X
ARCHETYPE
Pragmatist
A balanced, no-drama setup: some rules, some tools, nothing extreme.
CONTEXT TAX · EVERY TURN
~2.5k tokens
Moderate · median rig: 2.3k · breakdown
GUARDRAILS
2/5
Blocks destructive commands · No YOLO mode · details

Copy this rig

# review before running: this installs third-party code
$ npx degit aravindhsampath/agentic-template/.claude ./rig-agentic-template  # 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.

$ curl -fsSL --create-dirs -o .claude/agents/architect.md https://raw.githubusercontent.com/aravindhsampath/agentic-template/main/.claude/agents/architect.md
$ curl -fsSL --create-dirs -o .claude/agents/engineer.md https://raw.githubusercontent.com/aravindhsampath/agentic-template/main/.claude/agents/engineer.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 commit*--no-verify*)",
      "Bash(git commit*--amend*)",
      "Bash(git push*--force*)",
      "Bash(git push*-f*)",
      "Bash(git rebase:*)",
      "Bash(rm*.reviews*)",
      "Bash(rm*.git/hooks*)",
      "Edit(.reviews/**)",
      "Write(.reviews/**)"
    ]
  }
}

Subagents (2)

architectDesigns implementation strategy for non-trivial changes — features, refactors, migrations, design decisions, or anything touching multiple files. Reads the existing codebase, interrogates ambiguity, p
engineerImplements **exactly one** plan layer per invocation — builds it, commits, drives the per-commit review loop to approve (iterating with fix-up commits if reviewers request changes), reports, returns.

Permissions

deny (9)
Bash(git commit*--no-verify*)
Bash(git commit*--amend*)
Bash(git push*--force*)
Bash(git push*-f*)
Bash(git rebase:*)
Bash(rm*.reviews*)
Bash(rm*.git/hooks*)
Edit(.reviews/**)
Write(.reviews/**)
ask (0)
—
allow (21)
Bash(git status:*)
Bash(git diff:*)
Bash(git log:*)
Bash(git show:*)
Bash(git rev-parse:*)
Bash(git rev-list:*)
Bash(git branch:*)
Bash(git switch:*)
Bash(git checkout:*)
Bash(git add:*)
Bash(git commit:*)
Bash(git merge:*)
Bash(git pull:*)
Bash(git init:*)
Bash(git restore:*)
Bash(bash .claude/scripts/wait-for-reviews.sh:*)
Bash(bash .claude/scripts/open-reviews.sh:*)
Bash(bash .claude/scripts/branch-all-approved.sh:*)
Read(.reviews/**)
Read(.plans/**)
Read(references/**)

Similar rigs

copied ✓