~ / rigs / AliOcasha / claude_agents_skills_template

AliOcasha/claude_agents_skills_template

Claude Code project template: multi-phase agentic workflow with specialized agents, skills, rules and hooks

↗ GitHub ★ 0 Apache-2.0 updated 2mo ago project Claude Code

share on X

ARCHETYPE

Orchestrator

A bench of specialised subagents. The main agent mostly delegates.

CONTEXT TAX · EVERY TURN

~5.3k tokens

Moderate · median rig: 2.3k · breakdown

GUARDRAILS

3/5

Blocks destructive commands · Pre-tool screening hook · No YOLO mode · details

Copy this rig

# review before running: this installs third-party code
$ npx degit AliOcasha/claude_agents_skills_template/.claude ./rig-claude_agents_skills_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.

Skills (19)

architecture benchmark code-tour context-budget dev devops go-dev help python-dev qa requirements rust-dev setup skill-stocktake sql-dev sync-repos tester typescript-dev upgrade

Subagents (9)

Benchmark Engineer model: opus	Runs performance benchmarks against the {{STACK_TYPE}} stack, measures latency/throughput/concurrency, and reports regressions
DevOps Engineer model: opus	Manages Kubernetes manifests, Kustomize overlays, Helm charts, Helmfile environments, CI/CD pipelines, and deployment configuration. Use when implementing infrastructure features or fixes.
Go Developer model: opus	Implements Go features including HTTP handlers, controllers, adapters, middleware, and tests. Use after architecture is designed.
Python Developer model: opus	Implements and maintains Python data acquisition workers or scripts. Use after architecture is designed.
QA Engineer model: opus	Security audit, code quality review (YAGNI/KISS/CLEAN/SOLID), vulnerability check, and acceptance criteria validation. Reads docs/BUGS.md first to determine if this is a re-validation run.
Rust Developer model: opus	Implements core orchestration, computation, storage access, and gRPC/service logic in Rust. Use after architecture is designed.
SQL Developer model: opus	Implements PostgreSQL schema changes including tables, views, constraints, RLS policies, and migrations. Use after architecture is designed.
Test Engineer model: opus	Run unit, integration, and e2e tests. Implement e2e tests where needed. Close features when all tests pass.
TypeScript Developer model: opus	Implements TypeScript/React features and fixes frontend bugs. Use after architecture is designed.

Hooks (6)

event	matcher	runs
PostToolUse	Edit\|Write	${CLAUDE_PROJECT_DIR}/.claude/hooks/auto-format.sh
PreToolUse	Bash	${CLAUDE_PROJECT_DIR}/.claude/hooks/guard-destructive.sh
PreToolUse	Bash	${CLAUDE_PROJECT_DIR}/.claude/hooks/pre-commit-lint.sh
UserPromptSubmit	*	${CLAUDE_PROJECT_DIR}/.claude/hooks/bugs-reminder.sh
Stop	*	${CLAUDE_PROJECT_DIR}/.claude/hooks/stale-index-check.sh
Stop	*	${CLAUDE_PROJECT_DIR}/.claude/hooks/phase-driver.sh

Permissions

deny (9)

Bash(rm -rf*)

Bash(git reset --hard*)

Bash(git push --force*)

Bash(git push -f*)

Bash(git clean -f*)

Bash(git branch -D*)

Bash(*DROP TABLE*)

Bash(*DROP DATABASE*)

Bash(*TRUNCATE TABLE*)

ask (0)

—

allow (22)

Bash(git log*)

Bash(git diff*)

Bash(git status*)

Bash(git add*)

Bash(git commit*)

Bash(git ls-files*)

Bash(git show*)

Bash(git branch*)

Bash(git stash*)

Bash(git merge*)

Bash(git rebase*)

Bash(git fetch*)

Bash(git pull*)

Bash(ls*)

Bash(rg*)

Bash(grep*)

Bash(find*)

Bash(jq*)

Bash(tree*)

Bash(wc*)

Bash(head*)

Bash(tail*)

Context tax

Instructions: 4.0kMCP tool schemas (est.): 0Skill metadata: 897Subagent metadata: 404

ALWAYS-LOADED FILES (claude-code)

≡

CLAUDE.md · 73 lines · ~594 tok · 4 @import (+3.4k)

Estimate: characters ÷ 4 for instruction files and skill/subagent descriptions; MCP schema sizes from typical tool counts. Harnesses with deferred tool loading pay less. Method

Guardrails 3/5

✓

Blocks destructive commands

7 deny/ask rule(s) e.g. Bash(rm -rf*)

✗

Protects secrets

No deny rule for .env / keys / ~/.ssh

✓

Pre-tool screening hook

2 PreToolUse hook(s)

✓

No YOLO mode

Permission prompts stay on

✗

Sandbox or ask-first rules

No sandbox and no ask-first rules

Badge

Add to your README:

[![AgentRigs](https://agentrigs.dev/badge/rig/AliOcasha/claude_agents_skills_template.svg)](https://agentrigs.dev/r/AliOcasha/claude_agents_skills_template)

Indexed from commit 6bf699cd8166 on 2026-10-09. Data is structure only — env values and anything secret-looking are stripped.