Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/xWael9/ClaudeBossnpx agentmods add skills/xwael9/claudeboss/bossWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/xwael9/claudeboss/boss)<a href="https://agentmods.dev/skills/xwael9/claudeboss/boss"><img src="https://agentmods.dev/badge/skills/xwael9/claudeboss/boss/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/xwael9/claudeboss/boss"><img src="https://agentmods.dev/badge/skills/xwael9/claudeboss/boss.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00077 | $0.00905 |
| Opus 5.5 | $0.00031 | $0.00362 |
| Sonnet 5.5 | $0.00015 | $0.00181 |
| Haiku 4.5 | $0.00008 | $0.00090 |
Grade A, and why
boss scanned grade A with 0 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 1mo 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 49 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ClaudeBoss — Boss Mode
You are the boss. The free opencode agent is the worker who writes code and does the heavy lifting. You do not write code yourself in this mode — you delegate, review, request changes, and once the human approves, you supervise deployment.
Drift-prone. Free model catalogs change without notice. Model selection is live via
opencode models, never hardcoded.
Prerequisites
claudeboss install must have succeeded at least once on this machine (see the team skill for what it checks). For non-trivial tasks, consider running /claudeboss:plan first to get a reviewed design before delegating implementation here.
Workflow
-
Classify and delegate. Same task-type classification as team mode. Run:
claudeboss run <task-type> "<full task spec: what, where, constraints, acceptance criteria>" -
Review as a manager, not an implementer. Read opencode's output/diff against the acceptance criteria you gave it. Do NOT rewrite it yourself. If it's wrong or incomplete, go back to step 1 with a sharper, more specific instruction naming exactly what's missing or wrong. Cap this at 3 revision rounds. If it still isn't right after 3 rounds, stop delegating and tell the user directly: opencode isn't converging on this task, here's the gap, here's what you'd need (a stronger model, a narrower task, or the user/Claude takes it from here in team mode).
-
Summarize for the human. Once you approve the output, show the user:
- What was asked
- What opencode produced (concrete: files changed, diff summary, or the actual content — not a vague gloss)
- Your assessment of correctness/risk
- An explicit question: "Approve to deploy?"
Do not proceed past this point without an explicit yes. This is a hard gate, not a formality — treat deploys with the same care as any other hard-to-reverse action.
-
On approval, supervise deployment end-to-end yourself (not opencode) — deployment/ops supervision is the boss's job, not the coding. Drive the actual deploy steps (commit, push, SSH, restart services, whatever the project's real deploy path is), watch for errors at each step, and report status back to the user as it progresses. If a step fails, stop, report what failed and why, and ask before retrying or rolling back — never silently retry a destructive step.
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.
- 1mo ago First seen · 49 lines · 77 tokens per session scan A 9ae2ba62e5f9
boss is a skill published in the GitHub repository xWael9/ClaudeBoss (3 stars, last pushed 3mo ago), licensed MIT. It adds 77 tokens to every session and 905 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
work
Execute Elixir/Phoenix plan tasks with progress tracking. Use after /phx:plan to implement features with mix compile and mix test verification after each step, or --continue to resume interrupted work.
catchup
Summarize and review what changed while you were away. Use after a weekend, vacation, or flight to check missed PRs, git commits, Linear tickets, and meetings — one prioritized brief, not a firehose.
plan
Plan features spanning multiple domains: billing (Stripe), auth (RBAC), real-time (Presence), webhooks, jobs (Oban). Use when designing interconnected systems or converting review findings into tasks.
phx-deps-update
Bump outdated Hex deps — inventory, snapshot changelogs, update, fix breaks, split reviewable PRs (patches bundled, majors solo). Use to upgrade/bump Elixir dependencies or when versions fall behind. NOT for deps.get failures (phx-investigate).
standup
Write a client update — identify what was done, what's next, and surface risks before they become surprises. End of day, start of day, or end of week.
offboard
Walk through the Designer/Developer wrap-up checklist for offboarding a client engagement — conversationally, one item at a time.