Find plugins

by Thomas Kalka

Delegate research and small coding steps to cheap opencode subagents with the oc-sub tool, and watch and review each run.

by ManuOtel

Drive a self-hosted OpenCode instance as async background workers over MCP.

by ManuOtel

Drive a self-hosted OpenCode instance as async background workers over MCP.

by yangpeng

OpenEyes — AI-friendly computer-use primitives (capture/detect/click) as a CLI and MCP server. Drives any desktop GUI on Windows/macOS/Linux via the platform accessibility tree, with Vimium-style grid fallback.

by Guiderail

Australian FP&A pack on openfpa: 30 June years, Xero AU, GST/BAS. Upstream is Jeff Brines / Guiderail.

by Open Hivemind

Ship every finished turn of this thread to your team's Open Hivemind server, secrets scrubbed on the laptop, and search the shared history from inside the agent.

by Open Hivemind

Ship every finished turn of this session to your team's Open Hivemind server, secrets scrubbed on the laptop, and search the shared history from inside the agent.

by openIndu 社区

openIndu org-wide Agent behavior principles (11), 20 maintainer agents, organization Issue intake, tiered SDLC workflows, and a main-branch protection hook

by OpenTideHQ

27 Agent Skills for detection engineering — OpenTide TVM/DOM/MDR authoring, KQL/SPL/FQL, platform tribal knowledge, and defensive internals.

by Sharrmavishal

Portable operating method: session lifecycle agents (session-start/end), pre-ship code review, deploy with live verification, and production log health check. All agents use {{placeholders}} that adapt to your project.

by Mai Lam

Design how a team runs: the operating model itself, decision rights, governance forums, and OKRs. For teams with no existing process to inherit.

by Steven Biss

Turns a colleague's background material into a validated operating model capture sheet for the Operating Model Explorer, with the matching engine and validator bundled.

by OperStack

Five commands and two skills for content sites that have to rank and be cited: run the sixteen quality gates and fix in order, preflight a page before writing, check llms.txt against the corpus, submit URLs to IndexNow, and measure AI-referral visibility. A post-edit hook runs the fast gates on every content file you touch.

by BrightWay AI

Use a natural-language chief-of-staff front door, inspect Nucleus health, record agent metrics, and register schedules when the host supports them. Delegates CRM pipeline analysis and forecasting to the growth plugin when installed.

by Opsphere

DevOps and SRE intelligence for ChatGPT Codex — investigate incidents, search Datadog logs, diagnose CI/CD pipelines, check deployments, and operate across AWS, Kubernetes, GitHub, Jira, Sentry, Cloudflare, and more through a secure remote MCP gateway.

by dobachi

Mathematical optimization / decision engineering expert: formulate decisions as LP/MILP/convex/stochastic models, pick and drive a solver, handle uncertainty (stochastic & robust), and validate solutions with feasibility and sensitivity checks. For decision-support algorithms that go beyond descriptive statistics/ML.

by coral

Optimize GPT-5.6 prompts while preserving intent and surfacing instruction, information, and context conflicts.

by Andreas Asprou

Ask the Oracle (GPT-5.2 with deep reasoning) strategic technical questions. Read-only advisor for architecture, security, and complex debugging.

by Not Rest Inc.

RENAMED → install `notrest@notrest`. This entry exists only to tell existing installs where the harness went.

by 5umm3r

Orca-managed worktree workflow, coordination rules, and repository initialization for Codex, provided as skills. The orca-init skill installs the launch gate into each repository; the Codex sandbox flags and the repository's AGENTS.md remain the other guardrails.

by haboshi

開発中に見つけた課題を「チケット起票 → 別 Orca worktree で新セッション起動 → カード設定 → 報告」までワンアクションでスピンオフするワークフロースキル。Orca IDE（stablyai/orca）の公開 `orca` CLI を前提とし、既定はフルハンドオフ（dispatch --inject / worker_done 監視はしない）。ユーザーが「監視して」「完遂まで見て」と明示したときだけ opt-in の監督モードになり、dispatch --inject → check --wait のローリング待機 → 検収 → 撤収 → キャッチアップ報告まで親セッションが閉じる。検収後の撤収（未 push 確認 → completed → worktree 削除）も担当。「この課題スピンオフして」「切り出して別セッションで」「チケット切って別worktreeで進めて」「撤収して」で発動。

by Tim Schreiber

Big asks. Small tasks. Right-sized models. Turns any spec or prompt into small, mechanical tasks tagged with the cheapest model and effort that can do each one, then runs them through tiered worker subagents, in parallel wherever it's safe, verifying and committing every task.

by Josh

Run a goal end to end like a senior engineer: pick an approach and say why, do the work or delegate the parts worth delegating, prove it with evidence, and report plainly.

by claude-base

Running several agents at once: wave-orchestration is the law sheet for dispatch, lane conduct, gates under one build mutex and union merges, and herdr-driving is how a session drives the panes.