By settlemint
Hook-based agent enhancement and workflow automation for Claude Code. Enhances native Explore, Plan, and General-purpose agents with Rule of Five convergence patterns.
npx claudepluginhub settlemint/agent-marketplace --plugin flowEnhances the Explore agent with Rule of Five convergence. Enforces 5-angle exploration, progressive disclosure for token efficiency, MCP tool awareness, and confidence scoring.
Enhances the general-purpose agent with Rule of Five convergence. Enforces 30-second reality check, evidence-based completion, and explicit convergence signals.
Enhances the Plan agent with Rule of Five convergence. Enforces 5-pass refinement, parallelization mapping, merge wall detection, and evidence-based completion criteria.
Adds PR context awareness during regular work. Shows unresolved comments, guides resolution after fixes.
Enhances review agents with high-signal filtering and validation protocols. Enforces evidence-based findings, confidence scoring, and five-pass convergence.
Fix all unresolved PR review comments and CI failures. Resolves threads after pushing fixes.
Deep codebase analysis with progressive disclosure docs in .claude/docs/
SettleMint development tools for Claude Code.
Claude is smart. It has excellent built-in capabilities for planning, coding, and problem-solving. These plugins don't replace Claude's intelligence—they enhance it.
Our approach:
.claude/ for analysis, generating learnings that feed back into plugin improvementsAfter installing the plugins, the typical development flow is:
/plan → Use Claude's built-in plan mode
→ Review and accept the plan
→ Claude implements the changes
/pr → Create a pull request
/fix-pr-reviews → Address reviewer feedback
One-line setup (recommended):
curl -fsSL https://raw.githubusercontent.com/settlemint/agent-marketplace/main/setup.sh | bash
This installs dependencies, configures marketplaces, and installs the full plugin suite.
Interactive (inside Claude Code):
/plugin marketplace add settlemint/agent-marketplace
/plugin install devtools@settlemint
/plugin install flow@settlemint
Or add to your project's .claude/settings.json:
{
"extraKnownMarketplaces": {
"settlemint": {
"source": {
"source": "github",
"repo": "settlemint/agent-marketplace"
}
}
},
"enabledPlugins": {
"devtools@settlemint": true,
"flow@settlemint": true
}
}
Modern development tools with MCP-first skills. Uses Context7 for up-to-date library documentation.
Slash Commands:
| Command | Description |
|---|---|
/commit | Create a conventional commit |
/branch | Create a feature branch |
/pr | Create a pull request |
/push | Push with smart QA check |
/sync | Sync branch with main |
Skills:
| Skill | Domain | MCP Source |
|---|---|---|
react | React 19 components, Tailwind v4 | Context7 |
tanstack-start | Full-stack React framework | Context7 |
radix | Accessible UI primitives | Context7 |
shadcn | shadcn/ui components | Shadcn MCP |
api | oRPC API routes | OctoCode |
drizzle | Drizzle ORM, PostgreSQL | Context7 |
zod | Zod v4 schemas | Context7 |
viem | Ethereum blockchain client | Context7 |
solidity | Smart contracts, Foundry | OctoCode |
vitest | Unit testing | Context7 |
playwright | E2E testing | Playwright MCP |
thegraph | TheGraph subgraphs | OctoCode |
helm | Kubernetes Helm charts | OctoCode |
turbo | Turborepo builds | Context7 |
restate | Durable execution | Restate MCP |
better-auth | Authentication | OctoCode |
i18n | Internationalization (i18next) | Context7 |
motion | Animations (Motion/Framer) | Context7 |
recharts | Data visualization | OctoCode |
pino | Structured logging | OctoCode |
terraform | Infrastructure as code | - |
tdd-typescript | Test-driven development | - |
troubleshooting | Debug patterns | - |
agent-browser | Headless browser automation | - |
ast-grep | Mass rename/replace | - |
code-health | Code health audit | - |
codex-patterns | Deep code reasoning | Codex MCP |
design-principles | Linear/Notion/Stripe design | - |
react-best-practices | React/Next.js performance | - |
rule-of-five | Multi-pass convergence | - |
typescript-lsp | LSP code navigation | - |
vercel-design-guidelines | UI review | - |
Hook-based workflow automation that enhances Claude's native capabilities.
Slash Commands:
| Command | Description |
|---|---|
/fix-pr-reviews | Fix PR review comments and resolve threads |
Skills:
| Skill | Description |
|---|---|
init-enhanced | Deep codebase analysis with progressive disclosure docs |
fix-pr-reviews | Address reviewer feedback with educational responses |
Hooks:
The flow plugin uses hooks to enhance Claude's behavior without replacing it. Hooks fire on specific events and add context or trigger actions.
Configure in .claude/settings.json under the env key:
| Variable | Values | Default | Purpose |
|---|---|---|---|
FORCE_AUTOUPDATE_PLUGINS | "1" / "0" | Disabled | Force reinstall all plugins on startup |
Example:
{
"env": {
"FORCE_AUTOUPDATE_PLUGINS": "1"
}
}
MIT
The development-workflow plugin for Claude Code — 35 skills organized around a 6-phase workflow (Think → Review → Build → Ship → Maintain → Setup), 24 agents, and an interactive setup wizard for rules, modes, hooks, and MCP servers.
Executes bash commands
Hook triggers when Bash tool is used
Modifies files
Hook triggers on file write and edit operations
External network access
Connects to servers outside your machine
Share bugs, ideas, or general feedback.
Interactive setup wizard for configuring any repository with Claude Code best practices, based on Boris Cherny's workflow
Complete AI coding workflow system. Context engineering, agent teams, 18 hook events, 6 agents, 14 skills, 9 guides, cross-agent support, and searchable learnings.
Opinionated workflow guides and best practices - the preacher's proven patterns for Claude Code projects
Use this agent for optimizing human-agent collaboration workflows and analyzing workflow efficiency. This agent specializes in identifying bottlenecks, streamlining processes, and ensuring smooth handoffs between human creativity and AI assistance. Examples:\n\n<example>\nContext: Improving development workflow efficiency
Complete project development toolkit: 23 agents, 23 slash commands, 29 lifecycle hooks, and 69 reusable skills for Claude Code workflows