Workflow skills for Claude Code — standardized feature lifecycle: /discover → /define → /implement. Ships a family of lifecycle skills, shared protocols, an authoring standard, and the /new-skill scaffolder.
Parallel build worker for one work unit. Implements a single sub-issue or file group in the shared worktree. Spawned in parallel by implement-runner for multi-unit scope.
Codebase architecture scanner for /architecture. Reads existing patterns, module boundaries, and dependency graph to inform architecture decisions.
Architecture constraint analyzer for /architecture. Reads codebase-scanner and patterns-researcher outputs and maps system constraints, topology, integration risks, and assumption challenges.
Compound context analyst. Reviews session history and diff to extract what broke, what was tried, what worked, and why. One of three parallel compound extraction agents.
Architecture and design critique agent. Independently reviews architecture/design decisions for high-risk plans and identifies gaps, risks, and trade-offs.
Decide on technical architecture for a feature (components, data flow, APIs, dependencies).
Audit open GitHub issues for drift against repo state. Flags broken refs, stale claims, and contradictions.
Build a feature from a GitHub issue. Creates a git worktree and codes against acceptance criteria using TDD.
Context management protocol for rot reduction using editing, delegation, and summarization.
Capture learnings from completed work into durable wiki notes. Delegates to /save when agents-memo is available.
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Skill/agent collection for AI coding agents — a standardized lifecycle for feature development. Compatible with Claude Code and opencode.
%%{init: {'theme':'base', 'themeVariables': {'background':'#000','primaryColor':'#ffffff','primaryTextColor':'#000000','primaryBorderColor':'#000000','lineColor':'#000000','textColor':'#000000','titleColor':'#000000','clusterBkg':'#f3f4f6','clusterBorder':'#000000','edgeLabelBackground':'#ffffff'}}}%%
flowchart TB
subgraph Canvas["Skills flowchart"]
direction TB
Start([feature request]) --> Discovery
subgraph Discovery["/discover — explore & scope"]
direction LR
D_Describe["/describe — problem space"]
D_Specify["/specify — acceptance criteria"]
end
subgraph Define["/define — solution architecture"]
direction LR
Df_Arch["/architecture — tech decisions"]
Df_Design["/design — UX/UI"]
end
subgraph Implement["/implement — build loop"]
direction LR
I_Build["/build — TDD"]
I_Review["/review — correctness + standards"]
I_Verify["/verify — AC QA"]
I_Build --> I_Review --> I_Verify
I_Verify -.fix brief.-> I_Build
I_Review -.fix brief.-> I_Build
end
Discovery -->|create issue body| Define
Define -->|issue body| Implement
Implement --> PR([draft PR])
PR --> Compound["/compound — capture learnings"]
PR --> ResolvePR["/resolve-pr-feedback — batch review responses"]
PR --> WrapUp["/wrap-up — remove worktree + branch"]
subgraph Meta["Other tools"]
direction LR
EpicAutopilot["/epic-autopilot — autonomous epic→PR"]
IssueAutopilot["/issue-autopilot — single-issue end-to-end"]
NewSkill["/new-skill"]
FindSkills["/find-skills"]
Prune["/prune — audit memory"]
AuditIssues["/audit-issues — drift-check open issues"]
GrillMe["/grill-me"]
end
GrillMe -.-> D_Describe
GrillMe -.-> D_Specify
GrillMe -.-> Df_Arch
GrillMe -.-> Df_Design
GrillMe -.-> NewSkill
subgraph Obsidian["agents-memo (optional)"]
direction LR
Save["/save"]
WikiQuery["wiki-query"]
WikiLint["wiki-lint"]
end
Compound -. filing .-> Save
Df_Arch -. prior patterns .-> WikiQuery
Define -. prior decisions .-> WikiQuery
Prune -. vault audit .-> WikiLint
end
classDef canvas fill:#ffffff,stroke:#ffffff,color:#000000
classDef orch fill:#dddddd,stroke:#000000,stroke-width:2px,color:#000000
classDef spec fill:#eeeeee,stroke:#000000,stroke-width:2px,color:#000000
classDef prim fill:#a5b4fc,stroke:#000000,stroke-width:2px,color:#000000
classDef meta fill:#d1d5db,stroke:#000000,stroke-width:2px,color:#000000
classDef ext fill:#f9a8d4,stroke:#000000,stroke-width:2px,stroke-dasharray:4 2,color:#000000
class Canvas canvas
class Discovery,Define,Implement orch
class D_Describe,D_Specify,D_Scout,Df_Arch,Df_Design,I_Build,I_Review,I_Verify spec
class GrillMe,NewSkill,FindSkills,Prune,AuditIssues,WrapUp,Compound,ResolvePR,EpicAutopilot,IssueAutopilot,Save,WikiQuery,WikiLint meta
class Obsidian ext
Legend: phase orchestrators (gray subgraphs) spawn specialists (light nodes) that do the bounded work. Plugin-level tools run outside the phase lifecycle. The agents-memo subgraph shows integrations that activate only when installed.
claude plugin marketplace add misiekhardcore/agents-flow
claude plugin install agents-flow@agents-flow
Then enable it in your project or globally in Claude Code settings.
Add to your opencode.jsonc:
{
"skills": {
"paths": ["./skills"]
}
}
npx claudepluginhub misiekhardcore/agents-flow --plugin agents-flowObsidian based, agents knowledge memory. Sets up a persistent, compounding wiki vault with LLM-assisted ingestion, research, and retrieval. Covers memory management, session notetaking, knowledge organization, and agent context across projects.
Harness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Access thousands of AI prompts and skills directly in your AI coding assistant. Search prompts, discover skills, save your own, and improve prompts with AI.
Reliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques