~ / rigs / nferrer-dev / claude-dotfiles

nferrer-dev/claude-dotfiles

No description.

↗ GitHub ★ 0 no license updated 6mo ago personal setup Claude Code

share on X

ARCHETYPE

YOLO Cowboy

Permissions? Never heard of 'em. Ships at the speed of `--dangerously-skip-permissions`.

CONTEXT TAX · EVERY TURN

~3.6k tokens

Moderate · median rig: 2.3k · breakdown

GUARDRAILS

3/5

Blocks destructive commands · Protects secrets · Pre-tool screening hook · details

Copy this rig

# review before running: this installs third-party code
$ npx degit nferrer-dev/claude-dotfiles/agents ./rig-claude-dotfiles/agents
$ npx degit nferrer-dev/claude-dotfiles/commands ./rig-claude-dotfiles/commands
$ npx degit nferrer-dev/claude-dotfiles/skills ./rig-claude-dotfiles/skills
$ npx degit nferrer-dev/claude-dotfiles/hooks ./rig-claude-dotfiles/hooks

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 nferrer-dev/claude-dotfiles/skills/alpha-vantage .claude/skills/alpha-vantage
$ npx degit nferrer-dev/claude-dotfiles/skills/article-extractor .claude/skills/article-extractor
$ npx degit nferrer-dev/claude-dotfiles/skills/deep-research .claude/skills/deep-research
$ npx degit nferrer-dev/claude-dotfiles/skills/edgartools .claude/skills/edgartools
$ npx degit nferrer-dev/claude-dotfiles/skills/exploratory-data-analysis .claude/skills/exploratory-data-analysis
$ npx degit nferrer-dev/claude-dotfiles/skills/fred-economic-data .claude/skills/fred-economic-data
$ npx degit nferrer-dev/claude-dotfiles/skills/hedgefundmonitor .claude/skills/hedgefundmonitor
$ npx degit nferrer-dev/claude-dotfiles/skills/hugging-face-datasets .claude/skills/hugging-face-datasets
$ npx degit nferrer-dev/claude-dotfiles/skills/iterative-retrieval .claude/skills/iterative-retrieval
$ npx degit nferrer-dev/claude-dotfiles/skills/langsmith-fetch .claude/skills/langsmith-fetch
$ npx degit nferrer-dev/claude-dotfiles/skills/matplotlib .claude/skills/matplotlib
$ npx degit nferrer-dev/claude-dotfiles/skills/mcp-builder .claude/skills/mcp-builder
$ npx degit nferrer-dev/claude-dotfiles/skills/networkx .claude/skills/networkx
$ npx degit nferrer-dev/claude-dotfiles/skills/notebooklm .claude/skills/notebooklm
$ npx degit nferrer-dev/claude-dotfiles/skills/pdf .claude/skills/pdf
$ npx degit nferrer-dev/claude-dotfiles/skills/plotly .claude/skills/plotly
$ npx degit nferrer-dev/claude-dotfiles/skills/polars .claude/skills/polars
$ npx degit nferrer-dev/claude-dotfiles/skills/pymc .claude/skills/pymc
$ npx degit nferrer-dev/claude-dotfiles/skills/scikit-learn .claude/skills/scikit-learn
$ npx degit nferrer-dev/claude-dotfiles/skills/seaborn .claude/skills/seaborn
$ npx degit nferrer-dev/claude-dotfiles/skills/shap .claude/skills/shap
$ npx degit nferrer-dev/claude-dotfiles/skills/skill-creator .claude/skills/skill-creator
$ npx degit nferrer-dev/claude-dotfiles/skills/statistical-analysis .claude/skills/statistical-analysis
$ npx degit nferrer-dev/claude-dotfiles/skills/statsmodels .claude/skills/statsmodels
$ npx degit nferrer-dev/claude-dotfiles/skills/sympy .claude/skills/sympy
$ npx degit nferrer-dev/claude-dotfiles/skills/tapestry .claude/skills/tapestry
$ npx degit nferrer-dev/claude-dotfiles/skills/timesfm-forecasting .claude/skills/timesfm-forecasting
$ npx degit nferrer-dev/claude-dotfiles/skills/verification-loop .claude/skills/verification-loop
$ npx degit nferrer-dev/claude-dotfiles/skills/xlsx .claude/skills/xlsx
$ npx degit nferrer-dev/claude-dotfiles/skills/youtube-transcript .claude/skills/youtube-transcript

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(rm -rf *)",
      "Bash(rm -fr *)",
      "Bash(rm -r *)",
      "Bash(sudo *)",
      "Bash(mkfs *)",
      "Bash(dd *)",
      "Bash(shred *)",
      "Bash(truncate *)",
      "Bash(wget *|bash*)",
      "Bash(wget *| bash*)",
      "Bash(curl *|bash*)",
      "Bash(curl *| bash*)",
      "Bash(curl *|sh*)",
      "Bash(curl *| sh*)",
      "Bash(git push --force*)",
      "Bash(git push *--force*)",
      "Bash(git push -f*)",
      "Bash(git push *-f *)",
      "Bash(git push * -f)",
      "Bash(git reset --hard*)",
      "Bash(git clean -f*)",
      "Bash(git checkout -- *)",
      "Bash(git restore *)",
      "Bash(> /*)",
      "Bash(> ~/*)",
      "Bash(FORMAT *)",
      "Bash(format *)",
      "Bash(diskpart*)",
      "Bash(taskkill /F*)",
      "Bash(taskkill /f*)",
      "Bash(shutdown *)",
      "Bash(Restart-Computer*)",
      "Bash(Stop-Computer*)",
      "Bash(chmod 777 *)",
      "Bash(chmod -R 777 *)",
      "Bash(* /dev/sda*)",
      "Bash(* /dev/nvme*)",
      "Edit(~/.bashrc)",
      "Edit(~/.zshrc)",
      "Edit(~/.profile)",
      "Edit(~/.bash_profile)",
      "Edit(~/.ssh/**)",
      "Read(~/.ssh/**)",
      "Read(~/.gnupg/**)",
      "Read(~/.aws/**)",
      "Read(~/.azure/**)",
      "Read(~/.config/gh/**)",
      "Read(~/.git-credentials)",
      "Read(~/.docker/config.json)",
      "Read(~/.kube/**)",
      "Read(~/.npmrc)",
      "Read(~/.npm/**)",
      "Read(~/.pypirc)",
      "Read(~/.gem/credentials)",
      "Read(~/.claude/.credentials.json)"
    ]
  },
  "hooks": {
    "PreToolUse": [
      {
        "matcher": "Edit",
        "hooks": [
          {
            "type": "command",
            "command": "bash ~/.claude/hooks/config-protection.sh"
          }
        ]
      },
      {
        "matcher": "Write",
        "hooks": [
          {
            "type": "command",
            "command": "bash ~/.claude/hooks/config-protection.sh"
          }
        ]
      },
      {
        "matcher": "Bash",
        "hooks": [
          {
            "type": "command",
            "command": "CMD=$(jq -r '.tool_input.command'); if echo \"$CMD\" | grep -qiE '(^|;[[:space:]]*|&&[[:space:]]*|[|][|][[:space:]]*|[|][[:space:]]*)rm[[:space:]]' && echo \"$CMD\" | grep -qiE '(^|[[:space:]])-[a-zA-Z]*[rR]|--recursive' && echo \"$CMD\" | grep -"
          },
          {
            "type": "command",
            "command": "CMD=$(jq -r '.tool_input.command'); if echo \"$CMD\" | grep -qE 'git[[:space:]]+push.*(main|master)'; then echo 'BLOCKED: Use feature branches, not direct push to main' >&2; exit 2; fi"
          },
          {
            "type": "command",
            "command": "CMD=$(jq -r '.tool_input.command'); if echo \"$CMD\" | grep -qiE '(^|[;&|[:space:]])(kill[[:space:]]+-9|killall[[:space:]]|pkill[[:space:]])'; then echo 'BLOCKED: Process killing requires confirmation — run manually' >&2; exit 2; fi"
          },
          {
            "type": "command",
            "command": "CMD=$(jq -r '.tool_input.command'); if echo \"$CMD\" | grep -qiE '(^|[;&|[:space:]])(rmdir|unlink)[[:space:]]'; then echo 'BLOCKED: Directory/file removal requires confirmation' >&2; exit 2; fi"
          },
          {
            "type": "command",
            "command": "CMD=$(jq -r '.tool_input.command'); if echo \"$CMD\" | grep -qiE 'wsl[[:space:]]+--unregister'; then echo 'BLOCKED: WSL unregister is destructive — run manually' >&2; exit 2; fi"
          }
        ]
      }
    ]
  }
}

Skills (30)

Subagents (2)

general-code-reviewer model: opus	Expert code review specialist. Reviews code for quality, security, and maintainability. Use after writing or modifying code for a broad quality check (distinct from the plan-alignment code-reviewer in
security-reviewer model: opus	Security vulnerability detection and remediation specialist. Use PROACTIVELY after writing code that handles user input, authentication, API endpoints, or sensitive data. Flags secrets, SSRF, injectio

Hooks (16)

event	matcher	runs
UserPromptSubmit	*	bash ~/.claude/hooks/interface-guard.sh
UserPromptSubmit	*	bash -c 'set -o pipefail; jq -r .prompt \| PYTHONIOENCODING=utf-8 claude-trace-compactor --stdin \| jq -Rs "{\"updatedPrompt\": .}"'
PreToolUse	Edit	bash ~/.claude/hooks/config-protection.sh
PreToolUse	Write	bash ~/.claude/hooks/config-protection.sh
PreToolUse	Bash	CMD=$(jq -r '.tool_input.command'); if echo "$CMD" \| grep -qiE '(^\|;[[:space:]]\|&&[[:space:]]\|[\|][\|][[:space:]]\|[\|][[:space:]])rm[[:space:]]' && echo "$CMD" \| grep -qiE '(^\|[[:space:]])-[a-zA-Z]*[rR]\|--recursive' && echo "$CMD" \| grep -
PreToolUse	Bash	CMD=$(jq -r '.tool_input.command'); if echo "$CMD" \| grep -qE 'git[[:space:]]+push.*(main\|master)'; then echo 'BLOCKED: Use feature branches, not direct push to main' >&2; exit 2; fi
PreToolUse	Bash	CMD=$(jq -r '.tool_input.command'); if echo "$CMD" \| grep -qiE '(^\|[;&\|[:space:]])(kill[[:space:]]+-9\|killall[[:space:]]\|pkill[[:space:]])'; then echo 'BLOCKED: Process killing requires confirmation — run manually' >&2; exit 2; fi
PreToolUse	Bash	CMD=$(jq -r '.tool_input.command'); if echo "$CMD" \| grep -qiE '(^\|[;&\|[:space:]])(rmdir\|unlink)[[:space:]]'; then echo 'BLOCKED: Directory/file removal requires confirmation' >&2; exit 2; fi
PreToolUse	Bash	CMD=$(jq -r '.tool_input.command'); if echo "$CMD" \| grep -qiE 'wsl[[:space:]]+--unregister'; then echo 'BLOCKED: WSL unregister is destructive — run manually' >&2; exit 2; fi
PostToolUse	Bash	CMD=$(jq -r '.tool_input.command'); echo "$(date -Iseconds) $CMD" >> ~/.claude/bash-commands.log
PostToolUse	Bash	bash -c 'set -o pipefail; jq -r ".tool_response.stdout // \"\"" \| PYTHONIOENCODING=utf-8 claude-trace-compactor --stdin \| jq -Rs "{\"hookSpecificOutput\": {\"hookEventName\": \"PostToolUse\", \"updatedResponse\": {\"stdout\": .}}}"'
PostToolUse	Read	uv run ~/.claude/hooks/prompt-injection-defender/post-tool-defender.py
PostToolUse	WebFetch	uv run ~/.claude/hooks/prompt-injection-defender/post-tool-defender.py
PostToolUse	Bash	uv run ~/.claude/hooks/prompt-injection-defender/post-tool-defender.py
PostToolUse	Grep	uv run ~/.claude/hooks/prompt-injection-defender/post-tool-defender.py
PostToolUse	Task	uv run ~/.claude/hooks/prompt-injection-defender/post-tool-defender.py

Slash commands (4)

/desktop/orchestrate/telegram/update

Plugins (11)

superpowers@claude-plugins-officialcontext7@claude-plugins-officialpyright-lsp@claude-plugins-officialhookify@claude-plugins-officialmodern-python@trailofbitsproperty-based-testing@trailofbitsstatic-analysis@trailofbitsdx@ykdojosanctum@claude-night-marketclaude-notifications-go@claude-notifications-goflow-next@gmickel-claude-marketplace

Permissions

deny (55)

Bash(rm -rf *)

Bash(rm -fr *)

Bash(rm -r *)

Bash(sudo *)

Bash(mkfs *)

Bash(dd *)

Bash(shred *)

Bash(truncate *)

Bash(wget *|bash*)

Bash(wget *| bash*)

Bash(curl *|bash*)

Bash(curl *| bash*)

Bash(curl *|sh*)

Bash(curl *| sh*)

Bash(git push --force*)

Bash(git push *--force*)

Bash(git push -f*)

Bash(git push *-f *)

Bash(git push * -f)

Bash(git reset --hard*)

Bash(git clean -f*)

Bash(git checkout -- *)

Bash(git restore *)

Bash(> /*)

Bash(> ~/*)

Bash(FORMAT *)

Bash(format *)

Bash(diskpart*)

Bash(taskkill /F*)

Bash(taskkill /f*)

Bash(shutdown *)

Bash(Restart-Computer*)

Bash(Stop-Computer*)

Bash(chmod 777 *)

Bash(chmod -R 777 *)

Bash(* /dev/sda*)

Bash(* /dev/nvme*)

Edit(~/.bashrc)

Edit(~/.zshrc)

Edit(~/.profile)

Edit(~/.bash_profile)

Edit(~/.ssh/**)

Read(~/.ssh/**)

Read(~/.gnupg/**)

Read(~/.aws/**)

Read(~/.azure/**)

Read(~/.config/gh/**)

Read(~/.git-credentials)

Read(~/.docker/config.json)

Read(~/.kube/**)

Read(~/.npmrc)

Read(~/.npm/**)

Read(~/.pypirc)

Read(~/.gem/credentials)

Read(~/.claude/.credentials.json)

ask (0)

—

allow (30)

Bash(*)

Read(*)

Edit(*)

Write(*)

Glob(*)

Grep(*)

WebSearch(*)

WebFetch(*)

Agent(*)

NotebookEdit(*)

Skill(*)

TeamCreate(*)

TeamDelete(*)

SendMessage(*)

TaskCreate(*)

TaskGet(*)

TaskList(*)

TaskOutput(*)

TaskStop(*)

TaskUpdate(*)

mcp__*

EnterPlanMode

ExitPlanMode

EnterWorktree

ExitWorktree

CronCreate(*)

CronDelete(*)

CronList

ListMcpResourcesTool(*)

ReadMcpResourceTool(*)

Context tax

Instructions: 1.2kMCP tool schemas (est.): 0Skill metadata: 2.3kSubagent metadata: 141

ALWAYS-LOADED FILES (claude-code)

≡

CLAUDE.md · 100 lines · ~1.2k tok

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

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

✓

Protects secrets

Denies reads like Edit(~/.ssh/**)

✓

Pre-tool screening hook

7 PreToolUse hook(s)

✗

No YOLO mode

allows every Bash command

✗

Sandbox or ask-first rules

No sandbox and no ask-first rules

Badge

Add to your README:

[![AgentRigs](https://agentrigs.dev/badge/rig/nferrer-dev/claude-dotfiles.svg)](https://agentrigs.dev/r/nferrer-dev/claude-dotfiles)

Indexed from commit 3a4c6c7686c2 on 2026-10-09. Data is structure only — env values and anything secret-looking are stripped.

Similar rigs

CRAG666/dotfiles

Dotfiles Based on the catppuccin colors. Its Fancy

★ 47 setup Claude Code 29 skills 16d ago

Skill Collector 2.4k tok ·

Clay-HHK/claude-config

Personal Claude Code harness configuration: agents, commands, rules, hooks, skills, and docs

★ 2 setup Claude Code 78 skills 15 agents 7 hooks 4mo ago

YOLO Cowboy 12.3k tok ·

emoulson/dotfiles-claude

—

★ 0 setup Claude CodeCodex 13 skills 1 hooks 7mo ago

Skill Collector 17.3k tok ·

LeastBit/Claude_skills_zh-CN

Anthropics/skills中文学习版本来自：https://github.com/anthropics/skills

★ 589 project Claude Code 16 skills 9mo ago

Skill Collector 608 tok ·