qodeca/erfana
An agent-native, open-source Markdown workspace – run Claude Code and other terminal coding agents right beside your editor. macOS & Windows.
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
CONTEXT TAX · EVERY TURN
up to ~17.9k tokens
Claude Code loads MCP tools on demand, so the real cost is usually lower.
Context hog · median rig: 2.3k · breakdown
Copy this rig
# review before running: this installs third-party code $ claude mcp add chrome-devtools -- npx -y chrome-devtools-mcp@1.10.1 --isolated --headless $ npx degit qodeca/erfana/.claude ./rig-erfana # 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 chrome-devtools -- npx -y chrome-devtools-mcp@1.10.1 --isolated --headless $ npx degit qodeca/erfana/.claude/skills/releasing-erfana .claude/skills/releasing-erfana $ curl -fsSL --create-dirs -o .claude/agents/bug-investigator.md https://raw.githubusercontent.com/qodeca/erfana/develop/.claude/agents/bug-investigator.md $ curl -fsSL --create-dirs -o .claude/agents/code-implementer.md https://raw.githubusercontent.com/qodeca/erfana/develop/.claude/agents/code-implementer.md $ curl -fsSL --create-dirs -o .claude/agents/code-reviewer.md https://raw.githubusercontent.com/qodeca/erfana/develop/.claude/agents/code-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/codebase-explorer.md https://raw.githubusercontent.com/qodeca/erfana/develop/.claude/agents/codebase-explorer.md $ curl -fsSL --create-dirs -o .claude/agents/diff-summarizer.md https://raw.githubusercontent.com/qodeca/erfana/develop/.claude/agents/diff-summarizer.md $ curl -fsSL --create-dirs -o .claude/agents/docs-updater.md https://raw.githubusercontent.com/qodeca/erfana/develop/.claude/agents/docs-updater.md $ curl -fsSL --create-dirs -o .claude/agents/project-documenter.md https://raw.githubusercontent.com/qodeca/erfana/develop/.claude/agents/project-documenter.md $ curl -fsSL --create-dirs -o .claude/agents/refactoring-advisor.md https://raw.githubusercontent.com/qodeca/erfana/develop/.claude/agents/refactoring-advisor.md $ curl -fsSL --create-dirs -o .claude/agents/release-failure-analyzer.md https://raw.githubusercontent.com/qodeca/erfana/develop/.claude/agents/release-failure-analyzer.md $ curl -fsSL --create-dirs -o .claude/agents/release-notes-drafter.md https://raw.githubusercontent.com/qodeca/erfana/develop/.claude/agents/release-notes-drafter.md $ curl -fsSL --create-dirs -o .claude/agents/release-quality-runner.md https://raw.githubusercontent.com/qodeca/erfana/develop/.claude/agents/release-quality-runner.md $ curl -fsSL --create-dirs -o .claude/agents/security-auditor.md https://raw.githubusercontent.com/qodeca/erfana/develop/.claude/agents/security-auditor.md $ curl -fsSL --create-dirs -o .claude/agents/solution-architect.md https://raw.githubusercontent.com/qodeca/erfana/develop/.claude/agents/solution-architect.md $ curl -fsSL --create-dirs -o .claude/agents/test-writer.md https://raw.githubusercontent.com/qodeca/erfana/develop/.claude/agents/test-writer.md
This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 16.2k rigs:
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(./.env.*)",
"Read(**/.env)",
"Read(~/.ssh/**)",
"Bash(git push --force:*)",
"Read(**/*.pem)",
"Read(.env)",
"Read(~/.aws/**)",
"Read(**/*.key)",
"Read(**/.env.*)",
"Bash(git push --force*)",
"Bash(rm -rf:*)",
"Bash(rm -rf *)",
"Bash(sudo *)",
"Bash(sudo:*)",
"Bash(rm -rf /)",
"Bash(git push -f:*)",
"Read(.env.*)",
"Bash(git reset --hard*)",
"Bash(rm -rf /*)"
],
"ask": [
"Bash(git push:*)",
"Bash(git push *)",
"Bash(gh pr merge *)",
"Bash(git commit:*)",
"Bash(rm *)",
"Bash(gh pr merge:*)",
"Bash(gh pr create *)",
"Bash(git reset --hard:*)",
"Bash(npm publish:*)",
"Bash(git rebase *)"
]
}
} MCP servers (2)
| server | source | est. tokens |
|---|---|---|
| circuit-electron | local / custom | 2.5k |
| Chrome DevTools MCP · "chrome-devtools" | npm | 9.0k |
Skills (1)
Subagents (14)
| bug-investigator model: sonnet | MUST BE USED for investigating bugs and errors. Use when issue has 'bug' label or when diagnosing unexpected behavior. Performs root cause analysis and identifies fix locations. |
| code-implementer model: sonnet | MUST BE USED for writing production code. Use immediately after architecture approval. Implements features following approved plans and existing codebase patterns. |
| code-reviewer model: sonnet | Expert code review specialist. MUST BE USED before any git commit. Use immediately after code changes to review for security, quality, performance, and best practices. |
| codebase-explorer model: haiku | Fast codebase exploration specialist. Use PROACTIVELY in Discovery phase to find files, patterns, and affected code areas. MUST BE USED when searching for code locations, understanding project structu |
| diff-summarizer model: haiku | MUST BE USED for generating commit messages and PR descriptions. Use immediately before finalizing commits. Analyzes git changes and creates conventional commit messages following project standards. |
| docs-updater model: haiku | Simple documentation fixer. Use for Tier 1 issues involving typos, simple doc fixes, or minor text changes. Fast and lightweight for trivial documentation updates. |
| project-documenter model: sonnet | MUST BE USED for updating project documentation after features or before releases. Maintains CLAUDE.md, architecture docs, and changelog entries following established formats. |
| refactoring-advisor model: sonnet | Code refactoring advisor. Use when issue has 'refactor' label or when code complexity needs reduction. Identifies code smells and recommends improvements without implementing. |
| release-failure-analyzer model: sonnet | MUST BE USED when `release.yml` fails (Phase 3 of releasing-erfana skill, on non-zero `gh run watch --exit-status`) to analyse the failed run. Erfana-local agent that identifies which platform leg fai |
| release-notes-drafter model: sonnet | Erfana-local override. Produces two-tier release notes for Erfana at docs/release-notes/v{version}.md — a 3-5 bullet user-facing summary (supplied by the orchestrator after operator input) wrapping a |
| release-quality-runner model: sonnet | MUST BE USED in Phase 0 of the releasing-erfana skill to enforce the pre-flight checklist before any release tag is pushed. Erfana-local override. Validates: branch gate, clean working tree, required |
| security-auditor model: opus | Security review specialist. MUST BE USED for Tier 3 issues or when 'security' label present. Performs deep security analysis focusing on OWASP Top 10 and Electron-specific vulnerabilities. |
| solution-architect model: opus | MUST BE USED for system design and implementation planning. Use PROACTIVELY after Discovery phase before any implementation begins. Designs component structure, identifies risks, and creates detailed |
| test-writer model: sonnet | MUST BE USED when writing tests. Use PROACTIVELY after implementation or during TDD. Creates comprehensive unit and integration tests targeting >80% coverage for new code. |
Similar rigs
nylas/cli
Nylas CLI is a unified command-line tool for Nylas API authentication, email management (GPG signing/encryption), calendar, contacts, webhooks, timezone utilities, and OTP extraction
Fort Knox 1.8k tok · · Q49
Yassinekrn/claude-setup
Personal Claude Code setup — orchestrator/subagent delegation policy, custom agents, and a colored PowerShell statusline (git branch, model+effort, tokens, cost, rate limits).
Orchestrator 1.2k tok · · Q15
MCKRUZ/claude-code-mastery
The definitive Claude Code setup, configuration, and mastery skill — CLAUDE.md engineering, MCP servers, agent teams, hooks, CI/CD, and self-updating knowledge base
Minimalist 1.0k tok · · Q0
maryam-bahrami/claude-code-project-structure
A practical Claude Code project template showcasing best practices for CLAUDE.md, commands, skills, agents, MCP integrations, and team workflows.
Pragmatist 3.0k tok · · Q18