claude-mods

Heterogeneous cross-provider fleet - GLM (z.ai), Codex (OpenAI), Grok (xAI), Anthropic Sonnet/Opus/Haiku - from one session, porting the native Workflow tool's patterns (adversarial verify, judge panels, journal resume) to OS-process workers. Triggers: fleetflow, heterogeneous/mixed-model fleet, codex worker, grok worker, cross-provider fan-out, cross-model verify.

Landing discipline for parallel work: sequential test-gated landing queue, pre-land scrub, auto-rebase of in-flight lanes, fleet status, one-shot revert. Native primitives spawn; fleet-ops lands. Triggers: landing queue, land branches, merge queue, test gate, fleet status, land agent-team/background-agent branches, sequential merge.

One cheap headless Claude Code worker on a cheaper brain (GLM via z.ai, or Sonnet/Haiku) - full tool harness, isolated git worktree + CLAUDE_CONFIG_DIR - fanned out and gated by an Opus orchestrator. Triggers: fleet-worker, GLM worker, grunt worker, cheap parallel agent, offload to GLM, sonnet worker, haiku worker.

Router for parallel or recurring agent work across six skills. Covers: parallel agents, fan out work, delegate to workers, run overnight, scheduled loop, land branches, mixed-model fleet, orchestrate workers, background agents at scale. Triggers on: which skill for parallel work, fan out agents, spawn workers, run this overnight, schedule a loop, land my branches, heterogeneous fleet, delegate to cheaper model, autonomous loop.

Inter-session pmail - send and receive messages between Claude Code sessions running in different project directories. Uses global SQLite database at ~/.claude/pmail.db. Triggers on: mail, pmail, send message, check mail, inbox, inter-session, message another session, pigeon.

Security audit orchestrator - parallel dependency scanning, SAST pattern detection, auth/config review. Dispatches 3 audit agents simultaneously, consolidates into OWASP-mapped severity report. Triggers on: security review, security audit, OWASP, XSS, SQL injection, CSRF, authentication, authorization, secrets management, input validation, secure coding, vulnerability scan, dependency audit.

Evaluate current session for skill-worthy workflows and create reusable skills. Triggers on: auto-skill, create skill from session, save workflow, capture this as a skill.

Docker, Kubernetes, and AWS ECS/Fargate patterns. Triggers on: Dockerfile, docker-compose, kubernetes, k8s, helm, pod, deployment, service, ingress, container, image, ecs, fargate, task definition, ecs service, awsvpc, FARGATE_SPOT, ALB, ecs vs kubernetes.

Portless local-dev HTTPS proxy: replaces port numbers with named URLs (Caddy/nginx alternative for local dev). Triggers on: portless, local https proxy, named localhost URL, custom TLD, portless alias, portless.json, local CA trust, boot persistence, monorepo routing, Tailscale dev sharing.

FastAPI web framework patterns. Triggers on: fastapi, api endpoint, dependency injection, pydantic model, openapi, swagger, starlette, async api, rest api, uvicorn.

pytest testing patterns for Python. Triggers on: pytest, fixture, mark, parametrize, mock, conftest, test coverage, unit test, integration test, pytest.raises.

Cross-language testing strategies and patterns. Triggers on: test pyramid, unit test, integration test, e2e test, TDD, BDD, test coverage, mocking strategy, test doubles, test isolation.

Read and inspect LevelDB stores - especially Chromium/Electron app state (Local Storage, IndexedDB, Session Storage). Triggers on: leveldb, .ldb files, IndexedDB, Local Storage, Chromium storage, Electron app state, claude.ai cache, browser forensics, decode app state, claude desktop state.

Building applications ON Claude - the Anthropic API and Claude Agent SDK. Use for: anthropic api, claude api, messages api, tool use, function calling, prompt caching, agent sdk, claude-agent-sdk, structured output, json schema output, batches api, extended thinking, adaptive thinking, model selection, claude pricing, build claude agent, anthropic sdk, stop_reason handling, streaming claude, token counting, cache_control, output_config, tool_choice, agentic loop, rate limits anthropic.

Claude Code internals - hooks, skill frontmatter, headless mode, and extension debugging, current as of June 2026. Triggers: hooks, hook events, hook not firing, PreToolUse/PostToolUse/SessionStart, skill frontmatter, SKILL.md, skill not triggering, headless, claude -p, CI/CD scripting, plugin validate, MCP server not connecting, agent vs skill vs command vs hook.

Process Compose orchestration for non-containerized local services: process-compose.yaml schema, health checks, restart policies, dependencies, TUI/REST/MCP control, scheduling, and boot persistence. Use as a PM2/supervisord/Foreman replacement for local dev service management.

Terraform and OpenTofu infrastructure-as-code operations - project layout, state management, module design, plan/apply safety, CI/CD pipelines, and secrets. Use for: terraform, opentofu, infrastructure as code, IaC, tfstate, terraform state, terraform module, remote backend, terraform plan, terraform apply, for_each, moved block, terraform import, drift detection, tflint, checkov, HCL.

Pre-push safety gate for any git push to a remote (GitHub, GitLab, Bitbucket, self-hosted). Runs gitleaks + regex-layer secret scan, forbidden-file check, divergence check, size warning, and requires explicit confirm before pushing. Refuses on any secret hit. Triggers on: push to origin, push to github, push to remote, git push, can we push, safe to push, ready to push, pre-push check, push-gate.

Modern R for data analysis and statistics — tidyverse-first (dplyr, tidyr, ggplot2, native |> pipe), with base R and data.table as alternatives. Triggers on: R, Rstats, tidyverse, dplyr, tidyr, ggplot2, pivot/join/group, data.table, purrr map, broom, renv, Quarto.

Defend the agent against adversarial Unicode it obeys but a human can't see: bidi/Trojan Source, tag-char smuggling, zero-width text, homoglyphs. Triggers on: prompt injection, scan for hidden unicode, hidden characters, zero-width space, Trojan Source, ASCII smuggling, poisoned CLAUDE.md, MCP tool poisoning, is this file safe.

Author, index, and lint Architecture Decision Records — append-only memory that recovers the WHY behind a system's shape. Triggers on: adr, architecture decision record, decision log, record this decision, supersede an adr, why was this decided, adr template, next adr number.

ASUS router config and hardening: Asuswrt-Merlin, security hardening, encrypted DNS (DoT/DoH), VPN (WireGuard/OpenVPN), guest networks, VLAN/IoT isolation, AiMesh, AiProtection. Triggers on: asus router, asuswrt, merlin, wireguard router, AiProtection, AiMesh, nvram, jffs, IoT isolation.

Cloudflare Workers + Wrangler edge ops: runtime, bindings, local dev, secrets, deploy/CI, Pages-vs-Workers. Triggers on: cloudflare workers, wrangler, wrangler deploy, wrangler.toml, KV, D1, R2, durable objects, queues, vectorize, compatibility_date, edge functions.

Color for developers: color spaces, accessibility contrast, palette generation, CSS color functions, design tokens, dark mode. Triggers on: color, palette, contrast, WCAG, APCA, OKLCH, color-mix, dark mode colors, design tokens, CVD, color blind, P3, sRGB.

Comprehensive ffmpeg/ffprobe media processing: transcode, cut/trim/concat, color grading, loudness normalization, subtitles, GIFs, HLS packaging, hardware encoding, and quality gates (VMAF). Triggers on: ffmpeg, ffprobe, transcode, compress/convert video, extract audio, color grade, hls.

Git + worktree orchestrator: status survey, per-worktree triage, commits, PRs, branches, releases, rebases — reads inline, writes to a background agent. Triggers on: git status, anything to commit, anything to push, commit, push, create PR, rebase, release, tag, changelog, worktree, land worktrees.

GitHub remote operations: repo creation, metadata, releases, issue/PR management with preview-before-send, and read-only security auditing. Triggers on: push to github, ship release, gh release, audit github repo, gh issue, gh pr, merge PR, branch protection, secret scanning, SECURITY.md.

Create, refine, compose, and export isometric/dimetric assets for web and games: projection math, SVG/CSS/three.js generation, pixel-art and Blender workflows, engine tilemaps, ControlNet AI generation, and the iso-studio composer. Triggers on: isometric, dimetric, tileset, y-sort.

macOS workstation diagnostics: kernel panics, failing drives, launchd startup audit, wake reasons, TCC denials, APFS snapshots. Triggers on: mac is slow, won't boot, kernel panic, login loop, drive failing, launchd, TCC permissions, Full Disk Access, Gatekeeper, sysdiagnose.

Mapbox GL JS v3 (web) toolkit: custom markers, symbol/geojson layers, clustering, style expressions, 3D terrain/fill-extrusion, heatmaps, choropleths, camera animation, and performance gotchas. Use for any Mapbox GL JS work — markers, layers, style, 3D, or headless map verification.

Assess, validate, and adopt the Open Knowledge Format (OKF) across markdown+frontmatter knowledge bases. Triggers on: OKF, open knowledge format, knowledge bundle, frontmatter conformance, validate frontmatter, markdown knowledge base, adopt OKF, is this repo OKF-ready, index.md log.md.

Publish Python packages to PyPI via OIDC Trusted Publishing (PEP 740 attestations, gh-action-pypi-publish) instead of stored tokens. Use for: invalid-publisher errors, pending-publisher 404s, uv publish/twine, TestPyPI, environment approval gates, and rotating/auditing publish tokens.

Audit any repo against the agentic-quality doctrine — score entry docs, structure, and enforcement gates, then map each finding to its fix. Triggers on: repo doctor, repo audit, agentic quality, is this repo agent-friendly, doc drift, stale AGENTS.md, monorepo structure, nested CLAUDE.md.

Claude Desktop session toolbox: transfer sessions between accounts, recover an old session via a picker + AI handover brief, rebind cwd after a folder move, audit broken bindings, render an in-chat picker. Triggers on: summon, transfer/recover session, session picker, rebind, session doctor.

Behavioural-first defense against poisoned npm/PyPI packages in the publish-to-advisory window CVE tools miss. Use before every install or version bump: 7-day cooldown gate, Socket.dev behavioural score, stale-OIDC audit, publish-token rotation, and a worm-persistence self-scan.

Recolour, vectorise, and theme any SVG to a brand palette with a zero-dependency studio that emits a copy-paste CSS filter. Triggers on: recolour svg, brand tint an svg, duotone svg, recolour a diagram, cloudcraft/draw.io/figma svg to brand, svg css filter, png to svg, vectorise a logo, image trace.

Application/game-scale three.js: ES modules, GLTF pipeline (DRACO/KTX2/meshopt), AnimationMixer, physics (rapier/cannon-es), react-three-fiber, and performance at scale (InstancedMesh, LOD, draw calls). Triggers on: three.js, GLTFLoader, r3f, game loop, WebGL memory leak, boids.

yt-dlp media acquisition layer feeding ffmpeg-ops: format selection avoiding post-download transcodes, clip-at-download, cookies/auth, channel archive sync, SponsorBlock, subtitles, failure triage (403s, nsig). Triggers on: yt-dlp, download video/playlist/channel, youtube to mp3.

Generate PhD-level expert-agent prompt files (.md) for Claude Code - authoring, not runtime. For spawning parallel RUNTIME workers use fleet-worker/fleetflow (via parallel-ops). Triggers: spawn agent, create agent file, generate expert prompt, new agent persona.

Autonomous single-metric improvement loop - modify, measure, keep or discard, repeat (Karpathy autoresearch). One session, git as memory, bisect-on-regression. Triggers: iterate, improve autonomously, run overnight, keep improving, improvement loop, iterate until done.