#ModGitHub stars

1. 97floateeFrBmt-BIGetNouf/floatee-claude-modYour Floatee backlog inside Claude Code: the open todos above the prompt (or in a side pane), one button to process each, and guards that keep your API token out of the conversation1
2. 98lensPedroLaRosa/claude-codePuts tsc, linter and secret-scan diagnostics in the result of every Edit and Write, so Claude fixes what it broke in the same turn1
3. 99limit-guardburnrate-ai/burnrateWarns when the prompt cache is invalidated; near the 5h cap blocks subagents and wide scans and asks Claude for short answers.1
4. 100meadowDenisGuiraudet/claude-modsClaude's replies as pixel bubbles: a short sky and hill on top, a strip of grass under the last block1
5. 101next-boundary-guardMatheusChiodi/cockpitRefuses Next.js app router edits that break the client/server boundary: hooks without 'use client', or server-only code inside a client file1
6. 102nexusdarkomarijaan/nexus-modAll-in-one workspace suite: HUD and session monitor, prompt enhancer, assumption review, command shield, secret masking, turn checkpoints with rollback, visual design previews, auto-verify and clean delivery.1
7. 103notch-hudskuvshin0v/notch-hudA Dynamic Island-style HUD around the MacBook notch for Claude Code: session progress at a glance, answer questions and permission requests without going back to the terminal. macOS 15+ on Apple Silicon.1
8. 104read-guardLupusMalusDeviant/mod-boxHolds a whole-file Read (no offset/limit) of a file above the limit, a lockfile, minified code or a source map, and the same via cat, type or Get-Content in the shell; tells Claude the size, a token estimate and a better way (Grep, offset/limit). A second try at the same file goes through, and so does a Read with allow_large. Claude Code itself stops only at about 25,000 tokens.1
9. 105scope-copcandygoogy/notaryClaude stays inside the scope you set, and asks before leaving it. Edits outside it are flagged, or blocked if you want.1
10. 106secret-guardlegostin/claude-code-secret-guardKeeps secrets out of Claude's context: gitleaks scans every tool output, prompt and attachment before the model reads it, and you decide what the model sees.1
11. 107secret-maskFazzani/claude-modsMasks secrets (API keys, tokens, passwords, private keys) in the transcript; hover a masked value to reveal it.1
12. 108secretscrockalet/claude-modsHand 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
13. 109secretsfalkoro/claude-code-secretsClaude asks for API keys in a masked field above the prompt, saves them to the macOS Keychain, Linux keyring or Windows DPAPI, and uses them in Bash without ever seeing them1
14. 110smartworkbenchnathak/smart-workbenchSmartWorkbench for Claude Code: set the task contract, pin context, guard tool calls and verify the result1
15. 111test-guardJustmalhar/awesome-claude-modsBlocks edits that make tests pass by weakening them: new skip/only markers, fewer assertions, a gutted test file, or deleting a test file.1
16. 112trip-guardfhoces/learning-claude-codeCourse example mod: denies edits to the generated trips.csv, shows the last failed command in the status line, toasts when a turn runs long.1
17. 113tripwireMDmubarak786/claude-modsStop secrets leaving the machine. Holds an edit or command that writes an API key, a private key, or a credential literal, a command that reads a secret file and talks to the network, and a git commit whose staged changes contain a secret. Refuses by default.1
18. 114tripwireryx2/slopshopperHolds destructive commands until you say so: rm -rf, force pushes, git reset --hard, DROP TABLE, curl | sh, sudo, chmod -R 777, edits to .env. One question in Claude's own dialog, fail-closed if the guard itself breaks, and a /tripwire log of every decision.1
19. 115turn-receiptyoshihiko555/dotfilesKeeps a per-turn tally of edits, runs, curls and destructive commands and shows it on the turn footer; any non-read-only Bash counts as a run, a claim (English or Japanese) with no run flags the spinner, a destructive Bash unfolds its tool group, and an edit-without-run turn suggests running the tests.1
20. 116lemo-guardlemomo-ai/lemo-mod限时：Claude 单轮工作超时自动中止 · 等你确认的时间不计 · 默认关闭，在「安全」页开启 / Time limit: stops Claude when a turn runs over time · time waiting for you does not count · off by default, switch on in Safety54
21. 117cs-mermaidhex/claude-sessionsDraws mermaid flowcharts in replies inline, as coloured blocks and lines.52
22. 118harness-modsucsandman/Agnostic-AIHarness Mod layer: heartbeat/canary, native usage, rewrite explanations, measured subagent accounting, supervisor routing, secret redaction. Modes per guard in ~/.claude/mods/mods-config.json (classic | shadow_mod | mod).24
23. 119screen-guarddanyuchn/claude-modsMasks sensitive names, orgs and secrets in the transcript for screen sharing; /mask toggles, click a mask to reveal6
24. 120secret-redactorescapeboy/claude-code-kitHides tokens, keys and passwords in tool output as ‹secret:N›, restores them in later tool calls, blocks reading key files.97
25. 121scope-guardYeonwooSung/my-claude-code-modsScope Guard1.1k
26. 122rulebook-guardhamzafer/claude-code-modsWriting and git rules: replaces em dashes in prose, and asks before git commit --amend, an unformatted push, or personal info in notes and commits.426
27. 123ad-ldapseanGSISG/ad-ldap-pluginActive Directory administration over LDAPS for Claude Code: an MCP server (FastMCP + ldap3) with 15 tools, admin agents, slash commands, a safety skill, and a guard that refuses any AD write until the identical call has been dry-run first. Headline workflow: mass-assign computer managedBy from description-held owner names via a dry-run plan/review/apply loop.
28. 124adaflow-guardAdalink-ai/ada-modsGuardrail para apps integrados ao Adaflow: impede vazar o app token (x-ada-token) para o browser, para o codigo-fonte ou para o transcript
29. 125agent-cross-checknaoanao/agent-cross-checkWhen another coding agent (or a person) commits to your repo, Claude notices, runs mechanical checks (line endings, secrets, .env files, and the agent's report against the real files: missing files, silent videos, unconfirmed claims) and audits the commits by diff and tests. A git push waits until they are reviewed.
30. 126agent-guardAbyssCN/claude-lead-harnessGuard for dispatching subagents: denies isolation "worktree", checks a declared WORKTREE is a separate, clean, never-dispatched git worktree, and in opted-in repos requires non-read-only subagents to declare one.
31. 127agent-statekleinheinczg/claude-code-agent-state-modReports what a Claude Code session is really doing (idle / working / tool / permission, plus signals and token use) from inside the session, as a JSON snapshot and a JSONL event log
32. 128agent-stateM-i-k-e-l/agent-stateShows the state of Claude (working, permission, input, idle, background, compacting) in the iTerm2 tab subtitle and in a Sessions pane in the desktop app.
33. 129approve-blockedadammhal/claude-modsRetries an auto-mode-blocked call up to three times so Claude Code's own permission prompt (and Moshi) can ask you.
34. 130arcadeshabier/claude-pluginsFive games in the docked side pane: Rail Runner, Road Hopper, Block Drop, Flap, 2048. Hides below 110 columns.
35. 131archgateaarontan-bot/Claude-modArchitecture map + plan gate: the AI maps modules, declares a scope, you approve it, and edits outside that scope are blocked and observed.
36. 132arresterdot-agi/arresterAfter a hook, deny rule, or auto mode blocks a tool call, Arrester stops later calls that reach the same result another way or read or change your guards. After your No, it stops only exact retries.
37. 133assertion-guardianccdwyer/assertion-guardianBlocks edits that weaken tests to fake a green run: removed or loosened assertions, added skips, deleted cases, swallowed errors, snapshot rewrites
38. 134automodelmugenracerman/automodelAuto Model: picks the main model once per fresh conversation, locks it to protect the prompt cache, and hands small self-contained jobs to a helper.
39. 135autopilotJPanna/claude-modsWorks through a GOALS.md goal tree turn after turn, with locked checks, its own check verdicts, lessons that carry over, safety guards and a progress band
40. 136backlogliveweird/claude-modsA brief view of the repo's BACKLOG.md: what's next, and how many items are to-do, new, parked or blocked
41. 137bash-explainKKK12142/claude-modsExplains in Korean the shell command a permission dialog asks about, with what it would actually delete or discard (adapted from blast-radius), and asks before a risky loss in auto mode
42. 138bash-guardhugo88/claude-code-bash-guardBlocks catastrophic Bash commands; holds risky ones (rm -r, git reset --hard, git clean, force push, migrations) and shows what they would change, with Proceed and Cancel
43. 139blast-radiusaqaurius6666/claude-blast-radiusWhen a Bash command asks for permission, preview its blast radius: rm targets measured (files, size, what git can't restore), plus your own regex -> dry-run rules per command in a chain or pipe. A line under the dialog, details in a side pane. Display only: never changes the decision.
44. 140blast-radiuscskwork/claude-code-modsDry-runs risky Bash commands (recursive deletes, destructive git, kubectl delete, SQL wipes) and asks Proceed or Cancel first. Deletes inside system temp or .scratch/tmp/.tmp folders run without asking.
45. 141blast-radiusgauravruhela07/claude-modsShows how far each change can reach: the files that depend on every file Claude edits, and a flag on wide-reaching shell commands. Show only, never blocks.
46. 142block-battlejpo-oss/claude-gamesFalling-block puzzle inside Claude Code: solo Marathon, a bot to beat, 1v1 battles and a leaderboard while a long task runs
47. 143block-destructive-commandsJinhaoFang/claude-modsDenies Bash commands that match destructive patterns (recursive rm on roots, force push, hard reset, destructive SQL, disk formatting) before they run. A tool.call hook with a Bash matcher that returns { deny } without calling next.
48. 144block-explainerganbaroff/ai-guardrails-kitWhen one of this kit's guard hooks blocks a tool call, shows one plain line (what it means, what to do instead). Observe-only: never changes what the model reads. No files, no network. English by default, Russian by a one-line switch.