All Documents
3,528 documents available
agent
Presents a 12-step roadmap for building production-grade AI agents using LangGraph, covering design, deployment, and maintenance.
Guardrails - Safety & Compliance Layer
Documents a pluggable safety layer that intercepts LLM requests to rate-limit, block patterns, and redact PII.
Agent Guardrails
Defines guardrails for autonomous accounting agents covering prompt lifecycle, tool allow-lists, runtime enforcement, and incident response.
BandhuBol – Product Idea
Defines a multilingual AI companionship platform for emotional support, with freemium access and customizable avatars.
TinySDLC — soul.md
Defines the personality, values, and operating principles of a chat-native SDLC coach and build operator.
Splendor
Describes Splendor, a Rust-based kernel runtime that provides OS-like primitives for persistent, governable, neuro-symbolic agent loops.
Product Requirements Document: LLM-Guard
Specifies requirements for a Rust CLI tool that detects prompt injection and jailbreak attempts in LLM interactions with transparent risk scoring.
The AION Story
Chronicles the 9-month personal journey of building a 30-engine cognitive architecture for human-AI partnership, from isolation to open-source release.
ARCOS System — TODO (Programming-Oriented)
Maps a multi-component code generation system with 9 modules, each defined by entities, functionality, and acceptance criteria.
🌐 YOLO-Browser: Autonomous Web Automation Platform
Pitches an open-source autonomous web automation platform with trust-based permission mode, targeting a grant from the Block Goose ecosystem.
Testing Guardrails (Vitest)
Enforces 7 concrete testing rules for Vitest, covering mocking, timers, filesystem isolation, and teardown hygiene.
AWS Control Tower guardrails
Explains the two classification axes of AWS Control Tower guardrails: functional type (preventive vs detective) and enforcement level (mandatory, strongly recommended, elective).
Minimum guardrails for Government of Canada's cloud operationalization framework
Maps F5 BIG-IP hardening to 12 Government of Canada cloud guardrails with compliance notes and validation checklists.
AI Usage Best Practices Plan
Outlines a plan to use AI productively in development, with guardrails against duplication and over-reliance.
Part 5: Guardrails - Security, Privacy & Enterprise Governance
Teaches enterprise guardrails, security, privacy, governance, correctness, via a fictional decoder project with checklists and code examples in three languages.
MCP Use Cases
Explains how Model Context Providers inject live data from Jira, Confluence, Figma, GitHub, and browsers into AI prompts to improve coding accuracy.
Implementation Plan: Otterly Fun Swim School
Maps the full architecture, database schema, page routes, and phased implementation for a swim school booking platform built on Next.js and self-hosted Supabase.
Alice – emergente Monetarisierungs-Strategin
Defines a German-speaking AI persona that iterates through monetisation experiments using a c^c loop, audit gates, and an evidence ledger.
CLAUDE.md - Development Guide
Guides AI assistants on building and testing a Windows-only CLI tool that automates Classic Outlook via COM.
SPDX Tech Team Meeting 2023-01-03
Records minutes from an SPDX Tech Team meeting covering 2023 planning, quality SBOMs, serialization, and profile updates.
CLAUDE.md
Guides Claude Code to generate relevance judgments for product search queries using a three-phase agent architecture against the WANDS dataset.
README
Implements a trigger-aware lattice neural network for event detection as a sequence labeling task, reproducing the EMNLP 2019 paper.
🧪 LLM Integration Test Cases
Defines 34 test cases for an LLM-powered HR assistant covering greetings, policy queries, context, and edge cases.
LSS Search Quality Evaluation
Reports search quality metrics (NDCG, MRR, Recall, MAP) for BM25, embedding, and hybrid retrieval methods across synthetic and BEIR benchmarks.