AgentBar's Claude Code mod. Observes only: writes what Claude Code decided without asking, and what it measures (context, rate limits, running subagents), to ~/.agentbar/mods.d for the AgentBar app and CLI. Optionally draws one line above the prompt when another agent waits on you.
context-diet
drkokorev/context-diet
Trims huge tool outputs (logs, JSON, diffs, grep, file lists) before they reach Claude's context, and saves the full text to a file Claude can read
delete-guard
IT-Shelter-Labs/Claude-Mods
delete-guard
threads
IT-Shelter-Labs/Claude-Mods
threads
pinboard
sirkitree/pinboard
Keeps open decisions, the task list and links Claude creates in a sidebar pane, updated through its own tool
clawd-buddy
zhanbodev/clawd-buddy
Clawd, Claude Code's pixel mascot, lives above your prompt in the terminal and the desktop app: it acts out what Claude is doing, chats with you, dances to your music, plays a runner game, and in the desktop app idles in a little pixel San Francisco
fanyi
JrCx7scC/claude-code-fanyi
Write to Claude Code in your own language while Claude works in English: prompts are translated to English, and each reply gets a streamed translation underneath.
context-inspector
dwarvesf/dwarves-kit
Context waste analyzer for Claude Code: what fills the window, re-reads, near-duplicate outputs, what compaction kept, advice
loop-detective
dwarvesf/dwarves-kit
Detect repeated Bash failures and edit/retry cycles with evidence, live cases and investigation prompts
memory-lens
dwarvesf/dwarves-kit
Recalls the few memories related to each prompt: one line for the human, one context entry for the model. Local files only, no network on the hot path.
secrets-veil
dwarvesf/dwarves-kit
Masks secret values in tool results before the model reads them: named vendor variables, value-shape patterns, and high-entropy tokens. No reveal path.
Live panel of what Claude is doing: running tools with real commands and timers, what is waiting for your approval, subagents, and its current todo plan.
changes
mishgoldenberg/claude-mods
Every file Claude created or edited this session, with +/- line counts from git and one-click 'why?', summary and self-review.
guardrails
mishgoldenberg/claude-mods
Clickable safety rules with presets: block rm -rf, force-push, destructive git, secret files, sudo, installs, network; lock Claude to the project folder or read-only; add your own block patterns.
loop-breaker
mishgoldenberg/claude-mods
Notices when Claude is going in circles (the same command failing again and again, the same code rewritten over and over) and tells Claude to step back and rethink, and tells you too.
notify
mishgoldenberg/claude-mods
Notification inbox + native OS notifications when a long turn ends, a subagent or background task finishes, or Claude is stuck waiting for your approval.
toolbox
mishgoldenberg/claude-mods
Every tool Claude can use (built-in and MCP) in plain language, how often each was used, and suggestions tailored to the project you are in.
wire
rittmananalytics/wire-plugin
The Wire Framework — AI-accelerated data platform delivery for the full lifecycle: requirements, design, development, testing, deployment, enablement
spark-hud
Mel0nidas/spark-hud
An animated HUD above the Claude Code prompt: context window, plan limits (5h/7d) with burn-rate pace, cost, live subagents and your other Claude Code windows.
magician
Alexander-Tyagunov/magician
Skills and agents for the software lifecycle in Claude Code: research, design, planning, test-first building, code review, security checks and release. Includes a local code index, per-project workspace notes, stack-specific guidance at session start, and optional Jira and Confluence access.
dispatch-relay
Hangghost/learning-hacker-claude-mod
dispatch-board 的搭檔:worker session 把回報寫進結果檔後以一行結束,由 mod 在 turn 結束時把新增內容與最終回覆轉送給指揮站;背景工作進行中的回覆暫存、延後併送(mission 檔寫了 station 才啟用;需 Claude Code ≥ 2.1.296)
handoff-runner
Hangghost/learning-hacker-claude-mod
bg session 在 worktree 做完的分支,由你在面板按一顆按鈕快轉進 main:交接單只帶參數,步驟由 mod 推導,執行前檢查、執行後驗證