Tag: agentic-workflow
Automated GitHub Issue Resolution and PR Management
This skill automates the resolution of GitHub issues by orchestrating sub-agents to implement fixes and manage pull requests. It supports repository forking, continuous monitoring of PR reviews, and automated execution via watch or cron mod…
UnClick Persistent Memory Protocol
A protocol for managing persistent, cross-session memory using the UnClick MCP server. It enables agents to load, search, and save user preferences, identities, and session summaries to maintain context across interactions.
Structured development specification workflow
A workflow for managing medium-to-large changes through explicit requirements, technical design, and task planning. It ensures clear acceptance criteria and architectural alignment before implementation begins.
Automated GitHub Issue Duplicate Detection
This skill automates the process of identifying potential duplicate GitHub issues. It utilizes parallel agents and the gh CLI to search, summarize, and filter results, finally commenting on the original issue with a list of duplicates.
Generates expert-reviewed GitHub release notes
This skill analyzes all PRs merged between two version tags, cross-referencing linked issues and PR metadata. It classifies changes and dispatches specialized agents to assess breaking changes, producing a polished, publication-ready GitHub…
Automates User Story Implementation and PR Creation
This agentic skill takes a GitHub user story issue and autonomously manages the entire development lifecycle. It gathers context from RFCs, plans PR boundaries, writes code, generates tests, and submits the final pull request.
discover and plan fixes for flaky tests
This skill analyzes GitHub Actions failure logs to discover, rank, and diagnose flaky tests across a codebase. It orchestrates parallel agents to investigate root causes, map test coverage, and propose detailed, principled fix strategies.
Persistent Long-Term Memory Triage Protocol
An automated protocol for evaluating conversation turns to extract and persist durable, actionable facts using mem0. It employs a four-gate decision process to ensure information novelty, factuality, and security before committing to long-t…
CCG Latent Chain Mode
A structured workflow for executing complex, multi-step development tasks through a four-phase reasoning process. It optimises token usage by employing hidden-state reasoning and transmitting only context deltas.
Decision Intelligence for AI Agents
Enables agents to perform strategic decision analysis using graph-based dependency mapping and convergence scoring. It facilitates identifying critical paths, detecting information conflicts, and optimising choices through bandit and contex…
Exa Research Orchestrator
An orchestrator for executing deep research tasks by planning, dispatching subagents, and synthesising results using Exa searches.
adaptive research and implementation planning agent
This agent orchestrates complex development workflows through a structured pipeline: Understand, Research, Plan, Implement, and Verify. It synthesises evidence from local and external sources to generate actionable, robust implementation pl…
Automated Code Review and Auto-Fixing Skill
An agentic skill for performing deep code reviews, identifying bugs, and enforcing project-specific architectural patterns. It automatically resolves trivial issues like unused imports while providing a structured workflow for reporting and…
Independent Contract Compliance Verifier
This skill independently validates that a software implementation satisfies its defined contract by reading acceptance criteria, generating new test cases, and executing them using the project's existing toolchain. It serves as a secondary …
Deep Research MCP Server
An MCP server and Python API for executing deep research tasks using backends such as OpenAI, Gemini, and DR-Tulu. It supports agentic workflows featuring clarification loops, status polling, and multi-provider integration.
Structured feedback capture for agentic workflows
This tool allows developers to record structured feedback, including thumbs up/down signals, context, and rubric scores, following task completion or failures. It is designed to distill lessons from chat history and failed tool calls, feedi…
Record Lessons Learned in Memory
A procedure for documenting bug fixes, incidents, and discovered patterns into domain-specific memory files while checking for contradictions and stale entries.
Parallel Codebase Audit Orchestrator
Orchestrates parallel agent execution to perform codebase evaluations, health checks, and documentation audits. It organises the end-to-end scoping process and generates structured audit plans for pipeline integration.
Interactive Ralph Command Planning Assistant
An interactive assistant that guides developers through the collaborative creation of structured Ralph command sequences. It facilitates the definition of background context, environment setup, actionable tasks, and verification procedures.
Broad discovery and project orientation skill
This skill provides comprehensive orientation and broad discovery capabilities, ideal for developers unfamiliar with a repository. It orchestrates multiple specialized tools to help locate relevant code, schema definitions, project conventi…
Comprehensive agentic skill for code discovery
This skill provides comprehensive orientation and broad discovery capabilities, guiding users through unfamiliar repositories by coordinating specialized tools for searching, mapping, and contextualising code, schemas, and project conventio…
Critically review and refine agent outputs
This skill facilitates the critical evaluation and refinement of generated content, ensuring adherence to specified constraints and improving overall quality. It acts as a final validation layer for complex agentic workflows.
Secure API Proxy for Agentic Workflows
Janee acts as a secure, server-side proxy for external API calls, abstracting credential management. It encrypts and injects necessary authentication tokens, allowing agents to interact with services without ever handling raw API keys.
Ralph Autonomous Task Planning Skill
This skill enables work planning and the generation of task files for autonomous execution, including PRD creation and batch issue resolution.