~ / rigs / cesdperez / llmfiles

cesdperez/llmfiles

Dotfiles for LLM coding agents: Claude Code and Codex skills, commands, agents and hooks.

↗ GitHub ★ 2 no license updated 1d ago collection Claude CodeCodex
stack: Python
share on X
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
CONTEXT TAX · EVERY TURN
~3.1k tokens
Moderate · median rig: 2.3k · breakdown
GUARDRAILS
0/5
No committed guardrails · details
Quality 0/100 Guardrails v2

Copy this rig

# review before running: this installs third-party code
$ npx degit cesdperez/llmfiles/agents ./rig-llmfiles/agents
$ npx degit cesdperez/llmfiles/commands ./rig-llmfiles/commands
$ npx degit cesdperez/llmfiles/skills ./rig-llmfiles/skills
$ npx degit cesdperez/llmfiles/hooks ./rig-llmfiles/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 cesdperez/llmfiles/codex/skills/bugbounty .claude/skills/bugbounty
$ npx degit cesdperez/llmfiles/codex/skills/dailyworkrecap .claude/skills/dailyworkrecap
$ npx degit cesdperez/llmfiles/codex/skills/fleet .claude/skills/fleet
$ npx degit cesdperez/llmfiles/codex/skills/fullsend .claude/skills/fullsend
$ npx degit <redacted> .claude/skills/glablensedreview
$ npx degit cesdperez/llmfiles/codex/skills/glabsummary .claude/skills/glabsummary
$ npx degit <redacted> .claude/skills/locallensedreview
$ npx degit cesdperez/llmfiles/codex/skills/skillbuilder .claude/skills/skillbuilder
$ npx degit cesdperez/llmfiles/codex/skills/waitwhat .claude/skills/waitwhat
$ npx degit cesdperez/llmfiles/codex/skills/whichagents .claude/skills/whichagents
$ npx degit cesdperez/llmfiles/codex/skills/whichmodels .claude/skills/whichmodels
$ npx degit cesdperez/llmfiles/skills/glab-cli .claude/skills/glab-cli
$ npx degit cesdperez/llmfiles/skills/intervals-icu .claude/skills/intervals-icu
$ npx degit cesdperez/llmfiles/skills/jira-writing .claude/skills/jira-writing
$ npx degit cesdperez/llmfiles/skills/likec4 .claude/skills/likec4
$ npx degit cesdperez/llmfiles/skills/plainenglish .claude/skills/plainenglish
$ npx degit cesdperez/llmfiles/skills/supabase .claude/skills/supabase
$ npx degit cesdperez/llmfiles/skills/svelte .claude/skills/svelte
$ npx degit cesdperez/llmfiles/skills/zitadel-api .claude/skills/zitadel-api
$ curl -fsSL --create-dirs -o .claude/agents/Explore.md https://raw.githubusercontent.com/cesdperez/llmfiles/main/agents/Explore.md
$ curl -fsSL --create-dirs -o .claude/agents/codex-reviewer.md https://raw.githubusercontent.com/cesdperez/llmfiles/main/agents/codex-reviewer.md
$ curl -fsSL --create-dirs -o .claude/agents/codex-worker.md https://raw.githubusercontent.com/cesdperez/llmfiles/main/agents/codex-worker.md
$ curl -fsSL --create-dirs -o .claude/agents/general-purpose.md https://raw.githubusercontent.com/cesdperez/llmfiles/main/agents/general-purpose.md
$ curl -fsSL --create-dirs -o .claude/agents/gitlab.md https://raw.githubusercontent.com/cesdperez/llmfiles/main/agents/gitlab.md
$ curl -fsSL --create-dirs -o .claude/agents/reviewer-fast.md https://raw.githubusercontent.com/cesdperez/llmfiles/main/agents/reviewer-fast.md
$ curl -fsSL --create-dirs -o .claude/agents/reviewer.md https://raw.githubusercontent.com/cesdperez/llmfiles/main/agents/reviewer.md
$ curl -fsSL --create-dirs -o .claude/agents/worker-fast.md https://raw.githubusercontent.com/cesdperez/llmfiles/main/agents/worker-fast.md
$ curl -fsSL --create-dirs -o .claude/agents/worker.md https://raw.githubusercontent.com/cesdperez/llmfiles/main/agents/worker.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 *)"
    ]
  }
}

Skills (19)

Subagents (9)

Explore
model: sonnet
Read-only search agent for broad fan-out searches across many files, directories, or naming conventions when only the conclusion is needed. Replaces the built-in Explore so it runs on the newest Sonne
codex-reviewer
model: sonnet
Read-only audit handed to Codex, billed to the ChatGPT plan instead of Anthropic usage. Use for a review from a different model family, or to save Anthropic usage on an audit whose mandate stands on i
codex-worker
model: sonnet
Executor that hands a self-contained implementation, debugging, or research task to Codex, billed to the ChatGPT plan instead of Anthropic usage. Use when the brief stands on its own, needs no network
general-purpose
model: opus
General-purpose agent for multi-step tasks and research the other named agents do not fit. Replaces the built-in general-purpose so it runs on the newest Opus at high effort regardless of the session'
gitlab
model: sonnet
Creates GitLab merge requests using glab CLI. Use when the user wants to create an MR, especially in the background after implementing a feature or fix.
reviewer-fast
model: opus
Read-only audit agent for findings that hinge on pattern recognition against stated rules (test health, code standards, reuse, ticket alignment, performance smells). Runs on the newest Opus at medium
reviewer
model: opus
Read-only audit agent for findings that hinge on reasoning about execution (correctness, security, cross-repo impact, verification by running tests). Runs on the newest Opus at high effort regardless
worker-fast
model: sonnet
Executor for mechanical chores where the steps are known up front (renames, formatting, boilerplate, small config edits, running a documented command). Runs on the newest Sonnet at medium effort regar
worker
model: opus
Executor for delegated implementation, debugging, or research that needs full reasoning. Runs on the newest Opus at medium effort regardless of the session's own effort. Use for code changes, multi-fi

Slash commands (14)

/bugbounty/curatemds/dailyworkrecap/fleet/fullsend/glablensedreview/glabsummary/locallensedreview/repodigest/skillbuilder/teamdigest/waitwhat/whichagents/whichmodels

Similar rigs

copied ✓