AgentBBS cross-host agent federation for ruflo (ADR-164). Phase 1 exposes room coordination — federation_bbs_register / federation_bbs_publish / federation_bbs_watch / federation_bbs_human_join. Phase 2 (v3.40.0) adds signed cross-host federation — federation_bbs_identity (persistent Ed25519 node identity), federation_bbs_peer_add (pin a peer by nodeId + public key), federation_bbs_peers...
✦ rUv
ruflo-federation
ruvnet/ruflo
Cross-installation agent federation with zero-trust security, peer discovery, consensus-based task routing, and per-call budget circuit breaker (ADR-097)
✦ rUv
ruflo-ruvector
ruvnet/ruflo
Self-learning vector database via npx ruvector@0.3.13 — HNSW, adaptive LoRA embeddings, code-graph clustering, exact min-cut, RaBitQ/DiskANN extras, hooks routing, brain/SONA, 118 MCP tools. As a mod (ADR-445): a write guard that keeps secrets out of the vector store and shared brain, /ruvector-mod, and a status file the console shows
tavernweave-agent-skills
LiarMTTT/TavernWeave
TavernWeaveX无限月读 2.4.0: 23 shared skills for rolecards, software work and bounded cloud tasks, with capability checks and resumable handoffs; in Claude Code it injects the Host Front Door, adds the tavernweave agent and a Soul mode switch above the prompt. Actual execution depends on host tools and permissions.
brewmod-reminders
kochetkov-ma/claude-brewcode
Brewmod reminders -- durable file-based reminders (check prod tomorrow 09:00) that survive sessions: band at session start, offered at the first prompt, nothing runs unasked (Claude Code 2.1.301+ mod)
claude-jev
0x7067/claude-jev
Hands the small judgments in a session to TypeSafe's Jev, a System One model that returns typed answers instead of generating text: what kind of request each prompt is, which model tier a subagent needs, whether an edit breaks a rule already written in your CLAUDE.md or AGENTS.md, and which transcript blocks survive compaction. Kept blocks carry over verbatim — no LLM writes a summary.
verdict
ersinkoc/claude-mods
KOZMOS Verdict: test, build and lint results above the prompt. Spots the test runners, type checkers, linters and builds Claude runs (vitest, jest, pytest, go test, cargo, dotnet, mocha, bun, tsc, eslint, npm run build, maven, gradle…), parses pass/fail/skip and error counts, and shows chips like '✓ vitest 142 · ✗ jest 3 failed · ✓ tsc 0 errors' with a red pulse on failures and a sparkline of...
read-cap
VedantAndhale/claude-pro-kit
Stops Claude reading a file over 1,000 lines whole (with Read, cat or Get-Content) and gives it an outline tool (definitions with line numbers) so it reads only the range it needs. A retry still reads it whole.
jr-read-cap
jrlprost/claude-code-mods
Stops Claude from reading a long file whole: a Read of a file over 1,000 lines without a range is refused with a hint to grep first and read only the lines it needs. Saves context and tokens.
achievements
baselane-sh/mods
Unlocks eight badges once each (first session, first green test run, 100 and 1,000 tool calls, 7-day streak, 10 blocked calls, a 2 hour session, night owl), with a toast and /achievements.
bead-chime
baselane-sh/mods
Plays a short chime when a Bash call that ran bd close (or bd done) ends with exit code 0. It reads only the command text and the result, and runs no program.
bead-streak
baselane-sh/mods
Counts the days in a row you closed a bead, from Bash calls that ran bd close and exited 0. Adds /bead-streak and a toast on a new best. Reads command text only, runs no program.
session-clock
baselane-sh/mods
A band above the prompt with the wall clock and how long the session has run (14:05 · 1h 12m).
turn-timer
baselane-sh/mods
A band above the prompt with how long the current turn has run (turn 0:42). It shows while a turn runs and hides between turns.
systemap
0xfauzi/systemap
The systemap skill makes a system map from source facts and examined claims. Reader text obeys the configured writing profile: plain or ASD-STE100.
otto-hud
manuacl/claude-mods
Otto, a blue octopus, forecasts your context window on one line above the prompt, in the Claude Code desktop app, and in text in the terminal: weather scaled to auto-compact with what fills the context on hover, the tokens each recent prompt added, and your 5-hour and 7-day limits with a pace projection. Labels in English or French. Fork of Anthropic's Token Weather (Apache-2.0).
agent-compact-advisor
apolenkov/agent-compact-advisor
Status line score 0-100 for how good a moment it is to /compact now, with reasons; offers a ready /compact that keeps goal, decisions, leftovers and paths, and adds that template to every compaction. Never compacts by itself.
jev-scout
mako-code/gobworks
Jev ranks every file in the repo against each prompt in ~0.5 s and hands Claude the likely relevant ones (System-1 file triage).
model-router
Akurganow/model-router
Routes each subagent to a Claude model and effort that fits its task, judged by Jev. For each routed Agent call it sends the task text, condensed to at most 12,000 characters, the agent's description, cut to 500 characters, and the call's one-line description, all with token-shaped strings such as keys masked, to api.typesafe.ai, openrouter.ai or both, as its provider option allows, each with...
agent-pulse
afterever/claude-craftkits
A box above the prompt that always says what the agent is doing right now, alive with color and motion: an animated status glyph, a glowing jellyfish on the desktop, a timeline of the turn, and a history behind its triangle. Fork of Tzafrir Rehan's whats-agent-doing 0.4.3.
subagent-progress
afterever/claude-craftkits
Progress bar above the prompt plus an agents panel (/subagent-progress) for any subagent work: running, finished and planned subagents with model, task progress, context, cost and time, each drawn as a crab whose costume follows its model and effort. Fork of johnnyvizz/claude-kit savvy-progress 1.2.0.
waypoints
afterever/claude-craftkits
A route for /goal and any multi-step work, built from Claude's own task plan: the plan drawn as a line of nodes and legs above the prompt, an ETA at the goal's own pace, a pane that opens by itself with the whole task trail, and every chat's goal in /waypoints. Follows Agent-Pulse's palette and animation. Claude is asked to plan by the mod itself. Fork of Nate Herk's goal-meter 1.0.0.
highlighter
ParkerCai/highlighter
After each reply, Claude Haiku picks what to highlight: yellow for what needs your decision, blue for the key result, red for warnings, green for verified results. Color codes such as #0066CC are shown on their own color. The main model is never told. /highlight reads the last replies of a chat that has none; /highlight-preview shows what each color means.
read-guard
LupusMalusDeviant/mod-box
Holds 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.