Tag: agentic
Automated GitHub Issue Deduplication Skill
This skill automates the process of identifying potential duplicate GitHub issues. It utilizes parallel agents and the gh CLI to search, summarize, filter, and report up to three likely duplicates directly on the original issue.
Generate executable ADRs for agentic coding
This skill guides the creation of Architecture Decision Records (ADRs) that function as executable specifications for coding agents. It enforces a rigorous, multi-phase workflow, ensuring decisions are explicit, measurable, and include a de…
Foundational Agentic Skill Framework Implementation
This skill provides a foundational framework for implementing complex, multi-step agentic workflows. It allows developers to orchestrate various capabilities into cohesive, reliable processes.
Managing automated PR review cycles
This skill automates the entire PR review lifecycle, monitoring bot feedback, fetching inline comments, and guiding the developer through fixing real bugs and providing reasoned dismissals for speculative feedback. It ensures all bot commen…
Automated multi-agent code review for pull requests
This skill orchestrates a comprehensive, multi-agent code review process for pull requests. It systematically checks for bugs, logic errors, and adherence to internal compliance standards using specialized agents before posting validated, h…
Automated GitHub Issue Fixing and PR Workflow
This skill orchestrates the full lifecycle of fixing GitHub issues, fetching targets via the REST API and spawning sub-agents for implementation. It manages git operations, pushing code to a specified fork, opening pull requests, and monito…
Automated Pull Request Review Agent
This skill automates comprehensive Pull Request reviews, checking for logical correctness, security vulnerabilities, and adherence to architectural standards. It structures feedback using criticality badges and posts detailed inline comment…
Executable Architecture Decision Record Generation
This skill guides the creation of Architecture Decision Records (ADRs) that serve as executable specifications for coding agents. It enforces a rigorous, multi-phase workflow, ensuring decisions are measurable, self-contained, and include a…
File-based planning for complex agentic tasks
This skill implements a structured, file-based planning methodology, treating the filesystem as persistent working memory. It manages complex, multi-step tasks by maintaining state, findings, and progress across dedicated markdown files.
Scaffolding new Model Communication Protocol tools
This utility provides a comprehensive guide and boilerplate structure for developing new Model Communication Protocol (MCP) tools. It details the necessary steps, including class definition, service injection, and DI registration, ensuring …
Manual Context Compaction for Long Agentic Workflows
This skill suggests manual context compaction at logical breakpoints during extended, multi-phase tasks. Rather than relying on arbitrary auto-compaction, it prompts the user to explicitly compact the working memory when transitioning betwe…
General purpose agentic workflow skill framework
This skill provides a foundational framework for executing complex, multi-step agentic workflows. It enables structured orchestration of various tasks within the agentic pipeline.
Comprehensive Tech News Research and Aggregation Agent
This agent orchestrates parallel research across multiple tech domains (AI, DevTools, Security, etc.) by sweeping RSS feeds and cataloged sources. It validates findings using full-page content, deduplicates results, and generates a structur…
Automated PR Review and Comment Resolution Loop
This skill automates the entire PR lifecycle, starting with an initial code review and fixing findings, followed by committing and creating the PR. It then enters an iterative loop, fetching and automatically addressing all code review comm…
Persist conversation state into long-term memory
This skill persists the current conversation state into a dedicated memory store, ensuring critical facts, decisions, and architectural choices survive context compaction, session boundaries, or model handoffs. It provides structured method…
Comprehensive Codebase Technical Debt Auditor
This skill coordinates a comprehensive technical debt audit by scoping the codebase across four vectors: architectural, structural, operational, and hygiene. It executes a dedicated agent to generate a detailed, prioritized audit document s…
Orchestrates complex multi-step agentic workflows
This orchestrator manages the full lifecycle of complex agentic tasks, coordinating planning, parallel execution, validation, and learning. It ensures robust state management and dependency resolution across multiple worker agents.
Suppresses redundant praise and status updates
This skill acts as a suppression gate, preventing the agent from emitting redundant affirmations, praise, or status reports. It ensures that only novel or critical information is conveyed, thereby maintaining informational value and prevent…
TaskFlow pattern for inbox triage and routing
This skill demonstrates a robust pattern for managing inbox items by classifying them, routing them based on intent (e.g., waiting for a Slack reply or immediate notification), and persisting state across asynchronous steps. It allows the a…
CLI tool for deep LLM code review
This CLI tool facilitates deep, context-aware code analysis by bundling a prompt with selected project files for a single LLM pass. It supports multiple engines (browser/API) and manages long-running sessions for complex debugging or refact…
Automate GitHub Issues to Pull Requests
This skill automates the conversion of open GitHub issues into pull requests by fetching candidates, spawning background agents to implement fixes, and opening PRs. It also supports monitoring PR reviews and handling actionable comments to …
Multi-Model AI Debate and Synthesis Engine
This tool orchestrates structured, multi-round debates between various LLM providers, allowing for diverse perspectives on complex topics. It facilitates deep analysis and provides a final synthesis of the collective model viewpoints.
Automated GitHub PR Comment Review and Fix
Automates the identification, validation, and resolution of unresolved GitHub pull request comments. It uses sub-agents to apply surgical fixes to legitimate issues and provides automated replies to threads.
CMA Managed Agent MCP Server
This tool provides a managed agent server (CMA) for Claude, enabling tool use, code execution, and repository edits. It supports both local (stdio) and remote (HTTP) deployments, with a focus on simplifying agent interaction for non-technic…