peek

claude-pro-kit

Make the $20 Claude Pro plan last longer in Claude Code.

Twenty-two small mods that show you exactly where your usage goes and cut the waste. One mod makes a model call: prompt-polish, once each time you press Improve, and never on its own. The others make none. None adds anything to the system prompt (read-cap adds one tool, which Claude Code lists by name only until Claude first uses it; write-guard adds one line to a tool result at most once per conversation; compact-keeper adds a note of at most 4,000 characters after a compaction): every figure on screen is one Claude Code already reports, or a time the mod measured.

• With tool-diet, every request in a fresh session was 15,954 tokens smaller (−36%): 43,859 → 27,905, as the API reported. Details.
• With skill-diet on top of the other mods, every request in a fresh session was 8,226 tokens smaller (−30%): 27,423 → 19,197, as the API reported. Details.
• With agent-diet, a task that sends one Explore subagent cost 33% less on a Sonnet session: $0.05570 → $0.03741, averaged over three runs each. Details.
• In the benchmark, a debugging task cost 33% less with output-diet and reread-guard on, averaged over three runs each.

Install

In Claude Code:

/plugin marketplace add VedantAndhale/claude-pro-kit
/plugin install tool-diet@claude-pro-kit
/plugin install skill-diet@claude-pro-kit
/plugin install agent-diet@claude-pro-kit
/plugin install context-xray@claude-pro-kit
/plugin install pro-hud@claude-pro-kit
/plugin install output-diet@claude-pro-kit
/plugin install reread-guard@claude-pro-kit
/plugin install write-guard@claude-pro-kit
/plugin install loop-guard@claude-pro-kit
/plugin install cmd-diet@claude-pro-kit
/plugin install gh-account@claude-pro-kit
/plugin install cache-clock@claude-pro-kit
/plugin install read-cap@claude-pro-kit
/plugin install session-receipt@claude-pro-kit
/plugin install peek@claude-pro-kit
/plugin install budget-guard@claude-pro-kit
/plugin install turn-budget@claude-pro-kit
/plugin install compact-keeper@claude-pro-kit
/plugin install collision-guard@claude-pro-kit
/plugin install answer-pane@claude-pro-kit
/plugin install prompt-polish@claude-pro-kit
/plugin install kit-updates@claude-pro-kit

Updates are off by default for marketplaces you add yourself. With kit-updates installed you are told when a fix ships and /kit-update installs it, from the desktop app or the terminal. Without it, turn on auto-update once: in a terminal, run claude, then /plugin → Marketplaces → claude-pro-kit → Enable auto-update; the desktop app has no toggle for it.

Install any one on its own; they do not depend on each other. Mods are not sandboxed, so read the code before installing: each mod is a single file under plugins/<name>/hooks/.

tool-diet

Every tool listed in front sends its whole description and schema with every request. A deferred tool is listed by name only, and Claude loads it through ToolSearch when it needs it; Claude Code already does this for most MCP tools. tool-diet does it for the rest of the tools you are not using: anything outside the core set (Bash, PowerShell, Read, Edit, Write, Glob, Grep, Agent, Skill, ToolSearch, TodoWrite, AskUserQuestion) that you have not used in your last five sessions.

Measured with one prompt, "Reply with just OK.", in a fresh session, as the API reported each request:

On that setup, the largest tool moved was Artifact, whose description alone is 19,870 characters. What moves depends on your tools and your habits; /xray shows yours.

• The first time a deferred tool is used in a session costs one extra step, a ToolSearch call. After that it stays loaded for the session.
• Using a tool keeps it loaded for your next five sessions, so the set follows what you actually use.
• The choice is made once per tool per session, so the prompt cache is never disturbed mid-session. Changes apply from the next session.
• /tool-diet lists what is on demand this session, grouped by where each tool comes from; /tool-diet keep <tool> always loads one, unkeep undoes it, and /tool-diet off|on switches it. Answers are toasts, so they add nothing to the conversation. The status line keeps the count: 38 tools on demand.

skill-diet

Every request carries the skill listing: each installed skill's name and its whole description. With a few plugins installed that is dozens of skills, most of which a given project never uses (video skills in a backend repo, document skills in a game). skill-diet keeps the skills you used lately in this project listed in full and lists the rest on one line by name only. The Skill tool still loads any of them, and typing /name still works.

Measured with one prompt, "Reply with just OK.", in a fresh session with 45 skills installed and the other mods on, as the API reported the request:

In the same setup, asked to fill in a PDF form, Claude found anthropic-skills:pdf from its name alone and loaded it with the Skill tool. What moves depends on your skills; /xray shows yours.

• A skill used in this project in one of your last five sessions stays listed in full. Every use counts: typed as /name or called by Claude through the Skill tool.
• Usage is kept per project folder, so a skill you use in one repo does not crowd the listing in another.
• A skill installed after skill-diet stays listed in full for five sessions, so Claude can find it before you have used it.
• The listing is decided once per session, so the prompt cache is never disturbed mid-session. Changes apply from the next session.
• /skill-diet shows what is listed by name only this session and how many characters left the listing; /skill-diet keep <skill> always lists one in full, unkeep undoes it, and /skill-diet off|on switches it. Answers are toasts, so they add nothing to the conversation. The status line keeps the count, in the form <n> skills by name only, <n> characters off.
• /xray shows the skills line before and after, in tokens.

agent-diet

A subagent runs on your main model unless its definition or Claude's Agent call names another. Explore only searches and reads, yet in the author's own 85 subagent transcripts, every one of the 2,043 Explore requests ran on Opus or Sonnet, reading 147,298,992 input tokens. agent-diet starts the agent types you list (Explore by default) on Haiku.

• A model Claude names in the Agent call is kept, and a fork always runs on its parent's model.
• Other agent types (general-purpose, Plan, your own) keep their usual model.
• /agent-diet shows the setting and how many subagents it moved this session; /agent-diet model haiku|sonnet|opus picks the model, /agent-diet add|remove <agent type> changes the list, and /agent-diet off|on switches it. Answers are toasts, so they add nothing to the conversation. The status line keeps the count, in the form 2 subagents on haiku.

Measured with claude -p --model sonnet --output-format json on a task that sends one Explore subagent to find a file, three runs each in alternating order. Every run found the file. Figures are the cost and tokens Claude Code reported per model:

context-xray

/xray opens a pane with the exact breakdown /context computes: what is sent with every request (system prompt, tools, MCP tools, memory files, skills, messages), what is loaded on demand, which MCP tools load every time, and each memory file's size. It measures when the pane opens and when you press Refresh (or r), never in the background, because the exact count sends one token-count request per tool and memory file.

Opens by itself when the context crosses 60% and again at 80%, with a toast pointing at /compact and /handoff. /xray auto off keeps it to /xray only.

pro-hud

The recording at the top of this page is the band. In text:

Session    ━━━━━━━━━━━━━━──────   69%  resets in 32m
Week       ━━━━━━━━━━━━━━━─────   74%  resets in 4d 17h
Context    ━━━━━━━━────────────   44%  436,034 tokens
This turn  1m 35s · 1 tool · 0 files edited · 436,034 in · 431,260 cached · 580 out

On a wide window the three meters sit side by side on one row; on a narrow one each figure on the turn line wraps whole rather than being cut off.

• Session / Week: the share of your plan's 5-hour and 7-day allowance used, and when each resets. Account-wide: every chat and device counts.
• Context: how full this conversation is. Every request resends it, so a fuller context spends your session faster; /compact when it climbs.
• This turn / Last turn: updates live while Claude works (per tool call and per model request), then shows the turn's final figures and how many points of your session it used.
• The spinner gains · session 20%, and finished tool calls draw as one line: status dot, tool, target, time.
• A toast when the session crosses 80% and 90%.

/hud shows what is on; /hud all on|off, or /hud band|spinner|cards on|off. The answer is a toast, so toggling adds nothing to the conversation. It draws in the desktop app only and leaves the terminal as it is. Rows other mods put above the prompt still draw beneath the band.

Weekly toasts at 50%, 75% and 90% of the week, once each, because the weekly limit drains quietly across many sessions.

output-diet

When a Bash or PowerShell result runs past 120 lines or 8,000 characters, Claude reads:

[output-diet: 82/500 lines shown; all 500 in ~/.claude/projects/<project>/<session>/tool-results/output-diet-<call>.txt]

<first 30 lines>
… [lines 31–450 omitted; error/warning lines from them:]
   250│ ERROR: build failed in src/app.ts
   300│ Warning: deprecated API
…
<last 50 lines>

• You still see the output in the transcript as before; only Claude's copy is trimmed.
• The status line keeps a running count: 3 trimmed · 41,200 chars saved.
• The untrimmed text is written to disk first; if the write fails, nothing is trimmed.
• It goes in the session's own tool-results folder, beside the transcript, where Claude Code keeps the outputs it saves itself, so Claude can open it without a permission prompt. When that folder cannot be found, it goes under ~/.claude/output-diet/ (or CLAUDE_CONFIG_DIR).
• A Grep result past the same limits keeps its first 100 lines, with a note to narrow the search: [output-diet: first 100/252 lines shown; narrow the search, or read all in …]. In the author's 1,048 past Grep results, 26 went past the limits, and keeping the first 100 lines would have cut 104,052 characters. Glob is left alone: none of 128 went past.
• A subagent's report past 8,000 characters keeps its first 6,000 and last 2,000 characters, so the opening and the conclusion stay. In the author's 123 past reports, 17 went past it, 313,365 characters between them.
• Claude Code itself already cuts the middle out of very long shell output (past roughly 10,000 characters) before any mod sees it, and for a failing command no uncut copy is kept. output-diet works on what is left: the saved file holds everything Claude would have read, and an error line Claude Code cut is not in it.

reread-guard

When Claude asks to Read the same range of the same file again in the same conversation, and the file's size and modification time have not changed, the read is skipped and Claude is told to use the copy it has. Any edit, a different range, or a subagent (which has its own context) reads freely. Claude Code can clear old tool results from context, so retrying the identical read straight after a skip always goes through. The record resets on /compact and /clear.

What Claude is told in place of the repeat read, kept to one line because the model reads it:

api.ts unchanged since you read it; use that copy. If it's gone from context, retry the same Read.

Shell reads too. Claude often reads a file with the shell instead of Read: in the author's transcripts, 318 sed -n runs went past the guard. A shell command that only prints one file is now treated the same way: cat FILE, sed -n 'A,Bp' FILE, head -n N / tail -n N, and in PowerShell Get-Content, gc, cat or type (with -TotalCount, -Head, -First, -Tail, -Last or -Raw). The same range of the same unchanged file is skipped once, and retrying the same command goes through. A whole cat after a whole Read that showed every line counts as a repeat. A pipe, a chain, a variable, a glob or any other flag runs untouched. A shell read never counts as a Read, since Edit needs a real one first.

The status line counts them: 2 re-reads skipped.

write-guard

Write sends the whole file as Claude's output, the most expensive kind of token; Edit sends only the lines that change. Claude sometimes rewrites a file it has already read in full to change a few lines. In the author's own 272 session transcripts, 141 writes rewrote a file Claude had already read or written (999,273 characters), and 112 of them followed an earlier rewrite in the same session. Of the 512,847 characters in the rewrites whose previous version was in the transcript, 171,325 had changed.

A hook runs after Claude has written the content, so write-guard cannot save the rewrite it sees; it stops the ones after it:

• The first rewrite of an existing file in a conversation goes through, with one line for Claude after the result: You rewrote all of api.ts. For changes to an existing file use Edit: it sends only the changed lines.
• A later rewrite is held back once, and Claude is told: api.ts exists; change it with Edit, not a full Write. If a full rewrite is intended, retry the same Write. Retrying the same Write goes through.
• New files, and files under 2,048 bytes, are never touched. A subagent has its own conversation and its own first rewrite; a compaction or /clear starts over.
• The status line counts them: 1 full rewrite held back.

loop-guard

Each retry of a failing command re-sends the whole conversation, and the same command usually fails the same way. In the author's 272 session transcripts, 10 commands failed 3 or more times, 38 runs between them.

Once the exact same Bash or PowerShell command has failed twice in a row, the next try is held back once, and Claude is told:

This exact command failed 2 times in a row. Change the approach instead of rerunning it. If a rerun is intended, retry the same command.

• Retrying the same command straight after the hold goes through, for a command that is meant to be rerun.
• A success clears the count, and each command is counted on its own. A subagent counts its own commands; a compaction or /clear starts over.
• The status line counts them: 1 failing retry held back.

cmd-diet

output-diet trims long output after a command has run; cmd-diet keeps the noise from being printed at all. Before a Bash or PowerShell command runs, a known noisy command gets its own quiet flags, which drop progress lines and keep errors, test failures and warnings:

In a live headless run (git status && python -m pytest on 41 test files, one failing), the request cost 41,535 tokens without cmd-diet and 40,571 with it, by the API's usage, and both runs named the failing test and its reason correctly. The shorter output stays in context, so every later request in the session sends less too.

• Each step of a &&, || or ; chain is handled on its own. A step that pipes, redirects or substitutes (| grep, > file, $(...)) is left alone, since something else reads its output; 2>&1 is fine.
• A command that already sets its output level (git status --porcelain, pytest -v, curl -fsSL, npm install --silent) is left alone, and a flag already present is not added twice.
• If a tool rejects a flag (an old version), that rule stops for the session and Claude's retry runs the command as typed.
• The status line counts them: 3 commands quieted.

gh-account

With two GitHub accounts logged in to gh, a push to a repository the other account owns fails, and Claude spends turns on gh auth status and gh auth switch. Before a Bash or PowerShell command runs, gh-account matches each git push, pull, fetch, clone, ls-remote and gh step to the repository's owner, and the owner to a logged-in account. When that account is not gh's active one, that step alone runs with its token:

GH_TOKEN="$(gh auth token --user ACCOUNT)" git push

The token itself never appears in the command or its output, and the active account is not switched. For git it also asks gh for the credential.

• The owner comes from a URL in the command, -R owner/repo, a gh api repos/<owner>/... path, or the folder's remotes. With no remote named, every remote must point at the same owner.
• The account is the one whose login is the owner, else the one account that is a member of that organization. The accounts come from gh auth status, and organizations from gh api user/orgs. What it learns is kept across sessions.
• Nothing is guessed: no owner, more than one matching account, or anything unclear leaves the command as typed. It also leaves alone gh auth and other gh commands that reach no repository, a command that already sets GH_TOKEN, a token from the environment, git over ssh, hosts other than github.com, and subshells or substitutions.
• /gh-account shows which account each owner uses, and /gh-account forget clears it. Answers are toasts. The status line counts them: 2 commands sent as <account>.

cache-clock

Claude's prompt cache keeps your conversation for a fixed time after each request. Reply within it and the context is read from cache; reply after it and the whole context is sent again at full price. That message pays the cache-write price (1.25 times the input price for a 5-minute cache, 2 times for a 1-hour one) on every token instead of the cache-read price (0.1 times): 12.5 to 20 times more for the same context, and nothing on screen says so.

cache-clock puts the countdown in the status line, restarted by every response from the main conversation:

cache warm · 3m left
cache cold · next message re-sends 61,204 tokens

When the cache runs out, a toast says so once (after the lifetime is known, see below). The token figure is the previous response