Tag: llm-agent
Mechanical dispatcher for agentic build pipelines
This skill acts as the mechanical dispatcher for complex, multi-agent builds. It executes instructions received from an ephemeral orchestrator subagent, ensuring reliable state transitions and connecting various specialized components witho…
Generate structural design documents for systems
This skill generates a comprehensive structural design document, committing to module boundaries, data contracts, and system architecture before any implementation begins. It forces global decision-making and proactively audits the prompt f…
Agent Memory Migration Bridge for Enforcement
This CLI utility parses structured, per-role markdown memory files from the Agent Architect Kit and transforms them into actionable feedback entries for the ThumbGate runtime enforcement layer. It facilitates the migration of agent learning…
Multi-agent codebase evaluation and remediation planning
This skill orchestrates a comprehensive, multi-agent assessment of a codebase across twelve defined pillars (e.g., Architecture, Performance). It first scopes the evaluation via guided questions before running three parallel evaluators to g…
Refine Feature Ideas into Design Specifications
This skill guides the user through an interactive process of exploring a codebase to refine a raw feature concept into a comprehensive, structured design specification. It systematically gathers scope, architectural decisions, and data mode…
Trust infrastructure for AI agents and transactions
This comprehensive tool provides verifiable on-chain identity and trust scoring for AI agents, allowing developers to gate transactions and assess the risk of external agents or tokens. It supports agent registration, transaction guarding, …
Batch Create Structured GitHub Issues
This skill processes conversational brainstorming discussions, extracting raw ideas and decomposing large features into properly sized, actionable sub-issues. It then structures these into fully metadata-rich GitHub issues and commits them …
Autonomous Cortex Memory System Bootstrap Tool
This tool autonomously bootstraps the Cortex memory system, verifying infrastructure, building cognitive profiles, seeding the codebase, and importing historical memories. It executes a multi-phase setup, ensuring full system consolidation …
Initiate and manage voice calls programmatically
This tool enables agents to programmatically manage voice communications, supporting the initiation, continuation, and termination of calls. It integrates with major telephony providers including Twilio, Telnyx, and Plivo, while also offeri…
Advanced web search and content extraction tool
This suite provides structured tools for advanced web research, allowing developers to perform targeted searches with filters (e.g., topic, time range) or extract clean, chunked content from specific URLs. It supports both basic and advance…
Automated code review and quality assurance tool
This utility automates comprehensive code reviews using Codex or configured fallback LLMs, checking local, branch, or committed changes. It provides a structured closeout mechanism, integrating parallel testing and detailed reporting to val…
Persistent long-term memory for agent context
This skill provides advanced, persistent long-term memory for agents, categorising information into semantic facts, episodic events, and procedural workflows. It allows agents to recall user context, track history, and maintain state across…
Comprehensive Federal Campaign Finance Research Skill
This skill orchestrates multiple endpoints to perform deep-dive analysis of federal campaign finance data, covering candidate resolution, high-level financial posture, notable receipts, and compliance signals. It guides users through struct…
Comprehensive Project Documentation and Audit Tool
Alcove provides a structured framework for managing and auditing critical project documentation, including architecture, conventions, and technical debt. It allows developers to search, lint, and validate the completeness of project knowled…
UnClick Persistent Memory Protocol Implementation
This skill enforces a mandatory, four-step protocol for managing persistent, cross-session user context via the UnClick MCP server. It ensures the agent proactively loads, searches, saves facts, and saves session summaries to maintain conti…
Automated PR Review and Merge Readiness Handler
This skill automates the entire pull request review cycle, iteratively addressing all findings from human reviewers, AI bots, and CI checks until the code is merge-ready. It manages complex tasks such as paranoid pagination, cross-source fi…
Advanced Agent Memory and Context Management
This tool provides comprehensive memory lifecycle management for AI agents, enabling the assertion of facts with temporal metadata and salience scoring. It supports context window population, point-in-time temporal queries, and automated de…
GPT Researcher Autonomous Deep Research Agent
An autonomous agent that performs deep web and local research to generate detailed, cited reports using a planner-executor-publisher architecture. It supports custom retrievers, MCP data sources, and parallelised agent execution.
Automated Pull Request Draft Generator
Automatically generates structured pull request drafts including branch suggestions, titles, and descriptions by analysing git diffs and commit history. It identifies change types and assesses backward compatibility risks based on modified …
Building Model Context Protocol Servers for LLMs
This guide outlines the process for developing Model Context Protocol (MCP) servers, enabling sophisticated LLM interaction with external APIs. Developers can implement these servers using Python or TypeScript, ensuring robust tool definiti…