Celestia/Cosmos SDK development workflows with repo-aware context. 13 agents, 8 commands, 3 skills.
npx claudepluginhub celestiaorg/celestia-engineeringDocument solved problems to compound team knowledge for Cosmos SDK development
Enhance an existing plan with parallel research agents for depth, best practices, and implementation details
Create implementation plans for Cosmos SDK modules and features
Have multiple specialized agents review a plan in parallel before implementation
Resolve all PR review comments using parallel processing
Perform comprehensive code reviews for Cosmos SDK modules using specialized agents
Triage and categorize findings from code reviews, tests, or audits
Execute implementation plans for Cosmos SDK modules with quality gates
Use this agent when reviewing code for unnecessary complexity, over-engineering, and opportunities for simplification. This agent identifies convoluted logic, premature abstractions, redundant patterns, and code that could be clearer. Examples: - <example> Context: The user has implemented a new feature with multiple abstraction layers. user: "I've added the new delegation tracking system" assistant: "Let me have the code simplifier check for unnecessary complexity." <commentary> New features often introduce over-engineering that should be simplified. </commentary> </example> - <example> Context: The user has refactored existing code. user: "I've refactored the keeper to be more modular" assistant: "I'll use the code simplifier to ensure the refactor didn't add unnecessary complexity." <commentary> Refactors can sometimes trade one form of complexity for another. </commentary> </example>
Use this agent when reviewing code for gas efficiency, optimizing storage operations, and ensuring cost-effective transaction execution in Cosmos SDK modules. Examples: - <example> Context: The user has implemented a keeper method with multiple store operations. user: "I've added a batch update function for delegations" assistant: "Let me have the gas optimizer review the storage patterns." <commentary> Batch operations need careful gas optimization to avoid excessive costs. </commentary> </example> - <example> Context: The user is concerned about transaction costs. user: "Users are complaining about high gas fees for this operation" assistant: "I'll use the gas optimizer to identify optimization opportunities." <commentary> Gas optimization directly impacts user experience and chain economics. </commentary> </example>
Use this agent when reviewing IBC (Inter-Blockchain Communication) code, including packet handling, channel lifecycle, relayer interactions, and cross-chain token transfers. Examples: - <example> Context: The user has implemented IBC packet handling. user: "I've added the OnRecvPacket callback for our module" assistant: "Let me have the IBC expert review the packet handling logic." <commentary> IBC packet handling is complex and needs specialized review. </commentary> </example> - <example> Context: The user is implementing a custom IBC application. user: "I want to send custom data over IBC" assistant: "I'll use the IBC expert to review your IBC application design." <commentary> Custom IBC apps need careful design review. </commentary> </example>
Use this agent when reviewing proto/gRPC definitions in Cosmos SDK modules. Specializes in API design, backwards compatibility, and protobuf best practices. Examples: - <example> Context: The user has created new proto files for a module. user: "I've added the proto definitions for the new rewards module" assistant: "Let me have the proto expert review these definitions for API design and compatibility." <commentary> New proto files need expert review for backwards compatibility and API design. </commentary> </example> - <example> Context: The user is modifying existing proto messages. user: "I need to add a new field to MsgDelegate" assistant: "I'll have the proto expert check this change for backwards compatibility." <commentary> Proto message changes require careful review to avoid breaking existing clients. </commentary> </example>
Use this agent when reviewing Go code in Cosmos SDK modules. This agent applies best practices for proto definitions, keeper patterns, typed events, test conventions, and code simplicity learned from experienced Cosmos SDK reviewers (Damian Nolan, Rootul Patel). Examples: - <example> Context: The user has just implemented a new Cosmos SDK module. user: "I've added a new burn module to x/burn" assistant: "I've implemented the burn module. Now let me have the Cosmos SDK reviewer check it against best practices." <commentary> Since new Cosmos SDK module code was written, use the cosmos-sdk-reviewer agent to apply proto, keeper, and test conventions. </commentary> </example> - <example> Context: The user has modified keeper logic. user: "Please update the keeper to emit typed events" assistant: "I've updated the keeper to use typed events." <commentary> After modifying keeper code, use cosmos-sdk-reviewer to ensure it follows Cosmos SDK patterns. </commentary> assistant: "Let me have the Cosmos SDK reviewer verify the typed event implementation." </example>
Use this agent when reviewing Cosmos SDK code for security vulnerabilities. Specializes in IBC security, message validation, keeper permissions, and chain security patterns. Examples: - <example> Context: The user has implemented IBC functionality. user: "I've added IBC transfer support to our module" assistant: "Let me have the security auditor review this for IBC-specific vulnerabilities." <commentary> IBC code requires specialized security review for packet handling and channel security. </commentary> </example> - <example> Context: The user has created message handlers. user: "I've implemented the MsgBurn handler" assistant: "I'll have the security auditor check the message validation and authorization." <commentary> Message handlers need review for proper validation and authorization checks. </commentary> </example>
Use this agent when reviewing ABCI implementation, state transitions, determinism, and consensus-critical code in Cosmos SDK applications. Examples: - <example> Context: The user has modified BeginBlock/EndBlock logic. user: "I've added validator set updates in EndBlock" assistant: "Let me have the state machine reviewer check the ABCI implementation." <commentary> ABCI lifecycle methods are consensus-critical and need careful review. </commentary> </example> - <example> Context: The user is concerned about determinism. user: "Is my keeper logic deterministic?" assistant: "I'll use the state machine reviewer to audit for non-determinism." <commentary> Non-deterministic code causes consensus failures. </commentary> </example>
Use this agent when reviewing tests for Cosmos SDK modules. Specializes in integration tests, simapp setup, test coverage, and testing best practices. Examples: - <example> Context: The user has written tests for a new module. user: "I've added integration tests for the burn module" assistant: "Let me have the test reviewer check the test coverage and patterns." <commentary> New tests need review for coverage, proper setup, and following conventions. </commentary> </example> - <example> Context: The user is setting up test infrastructure. user: "I've created the test suite for the staking module" assistant: "I'll have the test reviewer verify the simapp setup and test structure." <commentary> Test infrastructure needs review for proper app setup and helper patterns. </commentary> </example>
Use this agent when reviewing chain upgrade plans, migration scripts, and version compatibility. Essential for safe mainnet upgrades. Examples: - <example> Context: The user is preparing a chain upgrade. user: "I need to upgrade from v1 to v2 with state migrations" assistant: "Let me have the upgrade validator review your upgrade plan." <commentary> Chain upgrades are high-risk and need thorough validation. </commentary> </example> - <example> Context: The user has written migration code. user: "Here's the migration for the new delegation format" assistant: "I'll use the upgrade validator to ensure the migration is safe." <commentary> State migrations can corrupt chain data if done incorrectly. </commentary> </example>
Use this agent when reviewing code that manipulates persistent state, database migrations, or any code affecting data integrity. Essential for state machine correctness and preventing data corruption. Examples: - <example> Context: The user has modified state management logic. user: "I've updated how we store validator sets" assistant: "Let me have the data integrity guardian review the state changes." <commentary> State management changes need careful review for data integrity. </commentary> </example> - <example> Context: The user is implementing a migration. user: "I need to migrate the old delegation format to the new one" assistant: "I'll use the data integrity guardian to ensure the migration is safe." <commentary> Data migrations are high-risk and need thorough review. </commentary> </example>
Use this agent when reviewing Hyperlane integration code, cross-chain messaging, ISM configuration, and interoperability implementations. Examples: - <example> Context: The user is integrating Hyperlane messaging. user: "I'm adding Hyperlane support to send messages to Ethereum" assistant: "Let me have the Hyperlane expert review your integration." <commentary> Cross-chain messaging needs careful security review. </commentary> </example> - <example> Context: The user is configuring ISM security. user: "I need to set up the Interchain Security Module" assistant: "I'll use the Hyperlane expert to review your ISM configuration." <commentary> ISM configuration directly impacts cross-chain security. </commentary> </example>
Use this agent when analyzing code for performance issues, optimizing algorithms, identifying bottlenecks, or ensuring scalability. Essential for consensus-critical code, state machine optimization, and gas efficiency. Examples: - <example> Context: The user has implemented a new query that iterates over state. user: "I've added a query to get all validators with their delegations" assistant: "Let me have the performance oracle analyze this for scalability." <commentary> Queries that iterate state can be O(n) and need optimization review. </commentary> </example> - <example> Context: The user is concerned about gas costs. user: "This transaction seems to use a lot of gas" assistant: "I'll use the performance oracle to identify optimization opportunities." <commentary> Gas optimization is critical for user experience and chain efficiency. </commentary> </example>
Use this agent when you need to perform security audits, vulnerability assessments, or security reviews of blockchain code. This includes checking for common vulnerabilities, validating input handling, reviewing authentication/authorization, and ensuring secure coding practices. Examples: - <example> Context: The user has implemented new message handlers. user: "I've added the MsgTransfer handler for cross-chain transfers" assistant: "Let me have the security sentinel audit this for vulnerabilities." <commentary> Cross-chain transfer code is high-risk and needs security review. </commentary> </example> - <example> Context: The user is concerned about input validation. user: "Can you check if our keeper properly validates all inputs?" assistant: "I'll use the security sentinel to audit input validation across the module." <commentary> Input validation is critical for blockchain security. </commentary> </example>
Cosmos SDK module development patterns and conventions for Celestia
Manage Git worktrees for isolated parallel development without branch switching
Update the celestia-engineering Claude Code plugin with latest changes from GitHub main branch
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.
Uses power tools
Uses Bash, Write, or Edit tools
The most comprehensive Claude Code plugin — 38 agents, 156 skills, 72 legacy command shims, selective install profiles, and production-ready hooks for TDD, security scanning, code review, and continuous learning
Manus-style persistent markdown files for planning, progress tracking, and knowledge storage. Works with Claude Code, Kiro, Clawd CLI, Gemini CLI, Cursor, Continue, and 16+ AI coding assistants. Now with Arabic, German, Spanish, and Chinese (Simplified & Traditional) support.
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
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.
Orchestrate multi-agent teams for parallel code review, hypothesis-driven debugging, and coordinated feature development using Claude Code's Agent Teams