bama
BAMA: Build Any Mac App. Agent skills to research, clone, design, build, optimize, and test native macOS apps with Swift and SwiftUI.
Duvo brings your people, AI agents and existing systems together to map, run and improve your processes, with decisions, exceptions and results in one place.
SkillsGitHub @duvoaiOnlineOwner unconfirmed
Have you tried it? One tap is enough. No sign-in needed.
No updates detected yet. GitHub releases will appear here.
Owners of public repositories under a personal GitHub account can sign in to verify ownership and manage the introduction themselves. No administrator approval is needed.
Sign in, generate a verification file, and commit it to this repository’s default branch within 30 minutes. Files on other branches or forks do not count. Then check the file below to manage the project without administrator approval. Keep the file to retain your access.
If needed, you will go to sign-in. After signing in, return here and press your chosen verification button again.
Run /nomorevibe verify in your project folder to verify ownership and manage the introduction yourself.
Verify in one minuteAsk about other verification options
Official agent skills for Duvo — automate operations with AI agents across your existing systems
Taken from the repository README
Loading the full README. The saved excerpt is shown below.
Official agent skills for Duvo — the AI platform that maps your real operations, designs the better version, and runs them across your existing systems.
These skills teach Claude and other coding agents to operate Duvo directly: manage Agents, debug failed Runs, and write the AOPs that drive them. How you install depends on which agent you use — see below.
There are three ways to install, depending on where you work:Read the full README ↗Official agent skills for Duvo — the AI platform that maps your real operations, designs the better version, and runs them across your existing systems.
These skills teach Claude and other coding agents to operate Duvo directly: manage Agents, debug failed Runs, and write the AOPs that drive them. How you install depends on which agent you use — see below.
There are three ways to install, depending on where you work:
skills.sh — installs the skills on their own, individually or as a set, paired with the duvo CLI.If you build and run your Duvo Agents from the Claude desktop app or claude.ai, install the Duvo plugin — it bundles the skills together with the Duvo connector (https://api.duvo.ai/v2/mcp), so a single install wires up both the know-how and the data access:
duvoai/skills.In Claude Code the equivalent is:
/plugin marketplace add duvoai/skills
/plugin install duvo@duvo
/reload-plugins
Notes:
duvo:run-debugger, duvo:aop-writer.duvo-cli — that one drives the terminal CLI, which doesn't apply inside the Claude app. Use the npx path below if you want it.This repository is also a spec-conformant Agent Plugin. Any Agent Plugins-capable client can install it straight from the repo and get all skills plus the Duvo MCP server in one go.
| Agent | How to install | Docs |
|---|---|---|
| Cursor | Customize in the sidebar → find the plugin → Install, choosing project or user scope. Cursor reads the format from the manifest, so an Agent Plugin installs like any other. To distribute it across a team, add duvoai/skills under Dashboard → Plugins → Add Marketplace → Import from Repo (Teams and Enterprise). | Cursor plugins |
| Hermes | hermes plugins install duvoai/skills --no-enable, then hermes plugins list and hermes plugins enable <plugin-name>. Portable packages stay disabled until you enable them explicitly. | Hermes plugins |
| OpenClaw | openclaw plugins install duvo --marketplace duvoai/skills. OpenClaw calls these bundles and maps them onto its native skills and MCP tools. | OpenClaw bundles |
| VS Code | Run Chat: Install Plugin From Source from the Command Palette and enter https://github.com/duvoai/skills. To pin it for a workspace instead, add "duvoai/skills" to chat.plugins.marketplaces in settings.json. | VS Code agent plugins |
| Claude app / Claude Code | Uses Claude's own manifest in .claude-plugin/ rather than the Agent Plugin path — see above. | Claude Code plugins |
| Anything else |
skills.shGet from zero to a working setup in four steps — install the skills, add the CLI they call, and sign in:
# 1. Install the skills into your agent
npx skills add duvoai/skills
# 2. Install the Duvo CLI (used by the skills to talk to Duvo)
npm install -g @duvoai/cli
# 3. Authenticate — opens your browser for OAuth sign-in
duvo login
# ...or sign in with an API key from https://app.duvo.ai/settings/api-keys
# duvo login --api-key dv_live_...
# 4. Verify you're signed in
duvo whoami
Then just ask your agent — for example: "Why did Run run_abc123 fail?", "Rewrite this AOP to be clearer", or "List my failed runs from yesterday".
npx skills add duvoai/skills
npx skills add duvoai/skills --skill duvo-cli
Skills installed via npx don't auto-update — run npx skills update to pull the latest versions.
Each skill is self-contained — install the whole set or just the one you need. All of them are available via npx; the Claude-app plugin ships all except duvo-cli.
| Skill | In plugin | Description |
|---|---|---|
| duvo-cli | — | Drive the Duvo platform from the terminal via the duvo CLI (@duvoai/cli) — manage agents, runs, cases, queues, files, skills, connections, and Clarity processes, or hit any endpoint with duvo api. |
| aop-writer | ✓ | Draft, rewrite, or critique a Duvo Agent AOP, returning a single Markdown document in the canonical GOAL / STEPS / NOTES shape. |
| run-debugger | ✓ | Investigate why a Duvo Run failed or produced the wrong outcome — reads the Run transcript and active Build, names the root cause from a fixed failure-mode taxonomy, and proposes one concrete fix. |
| workflow-debugger | ✓ | Audit a Duvo Agent — or a multi-Agent workflow connected by a Queue — across many Runs to find systemic inefficiencies and quality issues, then recommend concrete AOP and architecture changes. |
| connection-doctor | ✓ | Diagnose and health-check the Connections and credentials a Duvo Agent relies on — reads each Build's Connections and their live state, names the problem, and proposes one concrete fix. |
| improve-agent | ✓ | Run the guided, end-to-end loop that makes one Duvo Agent better — survey its setup, ground findings in recent Runs via the debugger skills, propose concrete changes, and apply them on your yes. |
MIT
No plugin support yet? Install the skills on their own via skills.sh — see below. Note that route carries no MCP server, so it needs the duvo CLI. |
| skills.sh |
| improve-queue |
| ✓ |
| Run the guided, end-to-end loop that makes one Duvo Queue — and the producer→consumer workflow around it — better, ending in applied AOP, topology, and trigger changes. |