Tools & agents

A pixel Luffy above the Claude Code prompt who acts out each tool call in five gears, an HP/MP/ST HUD for context, rate limits, and prompt cache, a shield that asks before destructive commands, a bug boss for failing tests, and a /pet stats pane. Your own pets live in a folder updates never touch.

The GTM operator mod for Claude Code: scores every go-to-market draft with its own skill pack's scorer, flags stale and inconsistent drafts across packs, pack dashboards, a guided sprint, analytics, and a guard that keeps brand-config.json and SOUL.md merged.

Measures a command: warmup and timed runs, median/p95/stddev and outliers, two commands compared or a saved baseline per project, and a pane with a block distribution plot. Bounded, one at a time, never during Claude's turn unless Claude asked.

Runs 2-4 attempts at a task in parallel git worktrees, each with a different approach; compare their diffs and adopt the best.

Finds the commit that broke something: git bisect with your test command in a temporary worktree (your working tree is never touched), live progress in a pane, the culprit with its diff stat handed to Claude.

Gives Claude a screen to draw on: SVG diagrams and Markdown reports in a Canvas pane with history.

Look into CSV, TSV, JSON and JSON Lines files without dumping them into context: schema, stats, rows, filters, group counts and single-row lookups, exact and compact, streamed with caps; a /peek pane with column sparklines.

An HTTP tool for API work: status, timing, trimmed headers and a body shaped to fit; localhost runs freely, other hosts ask once a session; an /http pane of the session's requests with replay and copy-as-curl.

A live plan Claude posts and keeps current: progress above the prompt, the full checklist and its history in a pane.

Search this project's past sessions: /recall for you (attach an old exchange to your next prompt), a tool for Claude to look up earlier decisions.

Runs your project's checks after Claude edits, shows pass/fail above the prompt, and closes the loop with one-press (or automatic, bounded) fixes.

Jev (TypeSafe's System One model) as Claude's instant co-processor: bulk filter/classify/rank/semantic-grep tools, risk notes under permission dialogs, a check on unverified 'it passes' claims, and optional per-prompt effort.

Wakes an idle Claude Code session when its local file mailbox receives an addressed instruction, and records the session's replies to files the sender can read. Independent and experimental.

A spellbook of agent skills for AI. Cast wisely. Four skills: eagle-eye, which lays coupled decisions out as a morphological box and renders one self-contained page that reads any configuration back; groundtrack, which turns a plan or a written change into a call graph with recorded traces through it, on a page a reader can step; contract, which interviews a person to agree the terms for a...

A Servers pane for this project's dev servers: what is listening, on which port, since when, with Start, Stop, Restart and Log buttons; names the process holding a port when a command fails with EADDRINUSE; a read-only dev_servers tool for Claude. Scans with netstat (never lsof), one session for all; no model calls.

Live pane of long-running jobs (training, downloads, extractions): progress, ETA, stalls and finishes, read from their logs with no model calls; jobs_list / jobs_watch / jobs_unwatch tools so Claude follows jobs instead of polling them.

Watches your open pull requests: CI pass/fail toasts for the PRs you open here, failing-check logs attached when you mention them, a prs_status tool so Claude reads them without gh, and the post-merge ritual (fetch --prune, switch, ff-only pull, delete the branch) done for you with no model calls; in a linked worktree it only fetches and leaves the worktree to you. Tells you once when gh's...

Makes past sessions recallable: indexes Claude Code and Codex transcripts, subagent runs, memory files, standing orders and second-opinion reviews into a local SQLite index (kept after Claude Code deletes old transcripts), gives Claude search, expand, recap and list tools, and adds /recall (search, last session, timeline, decisions, commands, files, PRs, ask), /remember, a...

A machine-wide meter of the plan's rolling 5-hour usage: every session's turns summed as effective tokens, the cap learned from each session-limit hit, the burn on the status line, a toast before the wall, /usage and a usage_status tool so Claude can pace itself. No process runs, no model calls.

A pane summarising the remaining todos, each with Do, Chat, Defer and Delete

/explore sends three subagents to investigate different parts of the codebase and merges their findings.

As tarefas do projeto num TASKS.md: painel, próxima tarefa acima do prompt, e o Claude criando e concluindo tarefas

Unity no Claude Code: erros de compilação no contexto, painel de testes, build do servidor e guardas de arquivos

A pane beside the session that Claude talks to you through: the task, what it is doing, the notes it files, the questions it went on without, and manual tests you run when you get to them.