Usage & cost
370 mods. Tokens, spend and rate limits at a glance.
- usage-probe 0.1.0 Writes the session's plan rate limits to the file named by THAI_SYLLABUS_USAGE_OUT 1
- usage-bell 0.1.0 Rings when the session nears a limit: context fill, rate-limit windows, the auto-memory index 1
- usage-limits 0.3.0 Keeps the 5-hour and weekly limit windows, their burn rate, a run-out projection, context fill, session cost and prompt-cache drops in a dim line above the prompt 1
- secret-shield 1.0.0 Keeps secrets out of Claude's context: API keys, tokens, private keys and passwords in tool output are replaced with [REDACTED:kind] before the model or the transcript ever sees them. 1
- context-bar 0.4.1 Stacked bar of the context window above or below the prompt, coloured by /context category, with rate limits, room before compaction and session length; /context-bar details for a full token breakdown 1
- auto-resume 0.1.0 When a turn dies on a rate limit (or an overloaded API), counts down to the reset and sends "continue" for you. 1
- usage-meter 0.1.0 Plan usage under the prompt: 5-hour, weekly and Fable limits with reset times, plus prompt-cache expiry 1
- agent-hub 0.1.0 Agent hub: a pane listing this session's subagents with status, model, turns and tokens, and a box to steer a running one. 1
- usage-limits 0.1.0 5-hour and weekly rate-limit bars in the prompt footer 1
- token-weather 0.5.1 A live forecast of the context window and the 5-hour and weekly quota, drawn above the prompt. 1
- secret-mask 0.2.0 Masks secrets (API keys, tokens, passwords, private keys) in the transcript; hover a masked value to reveal it. 1
- tools-usage 0.2.0 Side pane listing the tools used in the session (built-in, MCP, skill, agent, plugin, hook) with calls and token consumption. 1
- context-band 0.8.1 A band above the prompt with rate limits, tokens, speed, cache, cost and context fill, in light and dark themes 1
- cost-bar 0.1.0 A one-row band above the prompt with session cost, 5h/7d plan limits and an optional budget bar. 1
- secret-guard 0.1.0 Blocks Write, Edit and Bash calls that would put an API key, token or private key into a file or command. 1
- usage-widget 1.0.0 A bar per rate-limit window with time to reset and session cost. 1
- weather-widget 1.0.0 Context usage drawn as weather, from clear skies to a storm near compaction. 1
- stream-widget 1.0.0 A live gauge of how fast Claude is writing, in tokens per second, with a sparkline. 1
- sieve-widget 1.0.0 A compaction that keeps every word you and Claude said and folds only the old tool traffic, with no summary and no tokens spent 1
- ledger-widget 1.0.0 What a project has cost across every session, from any folder or worktree in it: measured as each turn ends, and read back from the sessions Claude Code saved 1
- attic-widget 1.0.0 Counts which tools Claude calls in this project, puts the unused ones behind ToolSearch, lists the daily ones up front, and reports how many schema tokens each request no longer carries 1
- secrets 0.2.0 Hand Claude an API token without it reading the value: you type it into a field above the prompt, Claude uses it in Bash as $NAME, and it is redacted from every tool result. 1
- compact-usage-meter 0.1.0 Context, session and weekly usage meters with reset countdowns on a row under the prompt hint line 1
- usage-limits 0.3.0 Keeps the 5-hour and weekly limit windows, their burn rate, a run-out projection, context fill, session cost and prompt-cache drops in a dim line above the prompt 1
- plan-meter 0.2.0 A band above the prompt with how far your plan file is (phases and steps done, what is in progress, what is next) and Claude's own task list, kept current as files change 1
- cost-guard 0.1.0 Watches session and daily spend, warns when the context grows large, and holds a prompt once when the prompt cache has expired on a big context. 1
- token-budget 1.1.1 Set a token or dollar budget per session; get warned at 80% and stopped at 100%. 1
- context-gauge 1.1.1 A slim band above the prompt showing how full the context window is. 1
- cache-hit-meter 1.1.1 Shows what share of input tokens came from the prompt cache. 1
- turn-timer 1.1.1 Times every turn and tells you when one runs long. 1