Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/destynova2/cli-code-skills/cli-forge-chefnpx skills add Destynova2/cli-code-skills --skill cli-forge-chefgit clone --depth 1 https://github.com/Destynova2/cli-code-skillsWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00175 | $0.13406 |
| Opus 5 | $0.00088 | $0.06703 |
| Sonnet 5 | $0.00035 | $0.02681 |
| Haiku 4.5 | $0.00017 | $0.01341 |
Grade C, and why
cli-forge-chef scanned grade C with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
| tmux | `which tmux` | `sudo dnf install tmux` | `brew install tmux` | `sudo apt install tmux` | None — required | Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
grep -q "CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS" ~/.claude/settings.json 2>/dev/null || echo "AGENT_TEAMS_NOT_ENABLED" How it starts
The opening of the file, as written. The whole thing — 929 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Optimization: Heavy content lives in
references/. Load on demand.
Language rule: Skill instructions are written in English. When generating user-facing files (prompts, shared-state, tmuxinator comments, reports), detect the project's primary language (from README, comments, docs, commit messages) and produce those files in that language. If the project is bilingual, ask the user which language to use before proceeding. The Brigade vocabulary (Menu, Commis, Sous-Chef, Chef, Plat, Mise en place, etc.) stays in French regardless — it's the pattern's canonical terminology, not prose.
Gotchas: Read
../gotchas.mdANDreferences/gotchas-chef.mdbefore producing output.
CLI Forge Boss — Brigade de Cuisine
"On ne cuisine pas seul. On cuisine en brigade." — Auguste Escoffier ("You don't cook alone. You cook in a brigade.")
The Brigade
┌─────────────────────────────────────────────────────┐
│ CLIENT │
│ (user, ticket, roadmap) │
│ │
│ Orders plats via the MENU (PERT) │
│ "I want: decision tokens + E2E + sokolsky" │
└──────────────────────┬──────────────────────────────┘
│ order
┌──────────────────────▼──────────────────────────────┐
│ CHEF DE CUISINE │
│ (plans, decides, orchestrates) │
│ │
│ - Reads the menu (PERT / roadmap) │
│ - Assigns plats to the commis │
│ - Decides the order in which to send them out │
│ - Asks the Sous-Chef to taste before sending │
│ - Calls "Envoyez !" when it's ready │
│ - NEVER COOKS │
└──────────────────────┬──────────────────────────────┘
│ SendMessage
┌──────────────────────▼──────────────────────────────┐
│ 3 VOTING SOUS-CHEFS │
│ (taste, judge — quorum 2/3) │
│ │
│ sous-chef-scope : is the plat in your station? │
│ sous-chef-secu : any broken glass in the plat? │
│ sous-chef-qualite : does it taste right? │
│ │
│ 2/3 APPROVE → passes │
│ 1 DENY → sent back to the kitchen │
│ 1 ESCALATE → call the patron (human, ~5% of cases) │
├──────────────────────────────────────────────────────┤
│ SOUS-CHEF MERGE │
│ (plates up, sends to the pass) │
│ │
│ - Quality gates (/cli-audit-*) │
│ - Merge + CI (tmux send-keys gate) │
│ - Updates the carnet (shared-state.md) │
│ - Runs in bypassPermissions │
│ - CWD = {project}-wt-gate │
│ (push + PR allowed, apply DENIED) │
├──────────────────────────────────────────────────────┤
│ APPLY PANE (only if infra present) │
│ (executes tofu/helm/kubectl after 3/3 quorum) │
│ │
│ - Receives APPLY_REQUEST from any agent │
│ - Runs the plan (tofu plan, helm diff, kubectl diff)│
│ - Submits plan to 3 voting sous-chefs — 3/3 needed │
│ - Executes only after unanimous APPROVE │
│ - Prod target → extra PATRON ACK │
│ - CWD = {project}-wt-apply │
│ (tofu/helm/kubectl allowed, git push DENIED) │
│ - See references/apply-quorum.md │
└──────────────────────┬──────────────────────────────┘
│ SendMessage
┌──────────────────────▼──────────────────────────────┐
│ COMMIS (x N) │
│ (cook, prep, plate) │
│ │
│ - Each has their own station (worktree / project) │
│ - Prepare their plat (code, test, commit) │
│ - Announce "Pret !" to the sous-chef │
│ - DO NOT DECIDE when to send out │
└──────────────────────┬──────────────────────────────┘
│ "Plat au passe"
┌──────────────────────▼──────────────────────────────┐
│ MAITRE D'HOTEL │
│ (pass → client, landing watchdog) │
│ │
│ - Receives PRs once auto-merge is enabled │
│ - Polls until MERGED (not just "auto-merge on") │
│ - Rattrapage: rebase BEHIND branches when main moves│
│ - Relance: rerun transient CI failures (max 2x) │
│ - Renvoi: real failures go back to the Sous-Chef │
│ - Encaissement: tag verified + shared-state updated │
│ - Reports "Client content" only when all 6 done │
│ │
│ Read references/maitre-dhotel.md for the full role │
└──────────────────────┬──────────────────────────────┘
│ "Client content + tag"
┌──────────────────────▼──────────────────────────────┐
│ CLIENT (happy) │
│ Release tag cut, prod deploy green, sprint done │
└─────────────────────────────────────────────────────┘
What ships with it
20 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- references/anti-patterns-chef.md 2.7 KB
- references/apply-quorum.md 9.3 KB
- references/brigade-setup-worktrees.sh 4.7 KB runs code
- references/ccheck-prompt-template.md 4.2 KB
- references/chef-prompt-template.md 38 KB
- references/conflict-resolution.md 10 KB
- references/contre-chef-inter-prompt-template.md 8.9 KB
- references/gotchas-chef.md 38 KB
- references/maitre-dhotel.md 14 KB
- references/parallel-exploration.md 9.1 KB
- references/permissions-template.md 17 KB
- references/pert-computation.md 17 KB
- references/quality-gates.md 1.9 KB
- references/raw-tmux-fallback.md 3.3 KB
- references/shared-state-template.md 8.4 KB
- references/simplified-model.md 9.8 KB
- references/sprint-persistence.md 7.0 KB
- references/sprint-report-template.md 8.0 KB
- references/templates.md 762 B
- references/tmuxinator-template.md 7.4 KB
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 929 lines · 175 tokens per session scan C 950365a8c90d
cli-forge-chef is a skill published in the GitHub repository Destynova2/cli-code-skills (5 stars, last pushed 9d ago), licensed MIT. It adds 175 tokens to every session and 13,406 once invoked, about $0.0009 per session on Opus 5. A static security scan graded it C with 2 findings (asks for root, reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
clean-architecture
Apply Robert C. Martin's Clean Architecture, SOLID, and component-decoupling principles when designing, reviewing, or refactoring code — scaled correctly to the size of the task so trivial changes don't get over-engineered. Use this whenever the user asks to design a new feature/service, review code for architectural…
sdlc
Full SDLC workflow for implementing features, fixing bugs, refactoring code, testing, releasing, publishing, and deploying. Use this skill when implementing, fixing, refactoring, testing, adding features, building new code, or releasing/publishing/deploying.
feedback
Submit feedback, bug reports, feature requests, or share SDLC patterns you've discovered. Privacy-first — always asks before scanning.
deploy
Deploy the application to production.
claude-setup-wizard
Setup wizard — scans codebase, builds confidence per data point, only asks what it can't figure out, generates SDLC files. Use for first-time setup or re-running setup.
claude-update-wizard
Smart update for SDLC wizard — shows changelog, compares files, lets you selectively adopt changes while preserving customizations.