ai agents
QUERY = 'contents'; CATEGORY = 'ai-agents' // 22 rows
Argus: Make Your AI Coding Assistant's Web Searches Visible
AI coding assistants run web searches during a conversation, but those searches stay invisible. Argus logs every WebSearch and WebFetch call to a local SQLite database, lets you query it from the CLI, and analyzes search patterns. I walk through what 68 days of real data revealed.
Compaction-Friendly Search Output: A Practical Playbook
There is a way to compress search output before the agent context window fills. file:line + 2 lines context format, file-cache, dedup, code examples for tool result wrappers, and measurement results.
Token Budget Arithmetic for Agent Search
An agent context window is a budget; search is a sub-budget. Where the ~15% search budget rule comes from, how it scales for different context window sizes, and how to catch overflow signals, with numerical examples.
Code Search for AI Agents: ripgrep, ast-grep, or Semantic?
ripgrep is fast, ast-grep is structural, semantic search is smart. But for an agent the right question is different: which backend in which order, under what token budget? Decision tree, academic evidence, and practical policy snippets.
ADR vs Spec-Driven Development: Why, What, and Using Both
Architecture Decision Records (ADR) and OpenSpec serve complementary roles: ADR answers 'why' a decision was made, spec defines 'what' gets built. How they work together in AI-assisted development.
From OpenSpec to Brainstorm + Court: Spec-Driven Workflow with Worktree
Evolution from OpenSpec CLI to a brainstorm + court pipeline, multi-AI evaluation, and parallel spec-driven development with git worktree.
Why KISS Is More Relevant Than Ever in the Age of AI Agents
According to Gartner, over 40% of agentic AI projects will be canceled. Why? Unnecessary complexity. Kelly Johnson's KISS principle, defined in the 1960s, plays a more critical role in AI agent architecture than in classical software.
Why SoC Is Critical in AI Agent Architecture
Separation of Concerns is not just a software principle but the fundamental design decision that determines whether an AI agent architecture collapses or stands. Context boundaries, defense layers, and protocol separation.
AI-Powered Codebase Audit: A Production-Grade Approach for Solo Entrepreneurs
Enterprise-quality codebase audit process using AI tools as a solo entrepreneur. Guardrail-driven AI-assisted coding, 6 audit tracks, and the Augmented CTO model.
Living Architecture Documentation for AI Coding Agents: Research, Approaches, and Tools
Codified Context, AGENTS.md ecosystem, C4 Model, Repomix, ADRs, Spec-Driven Development, Google Code Wiki. A comparative analysis of 11 different approaches to giving AI agents architectural context, backed by research.
Hybrid Search: Smart Search Architecture with FTS5 + Vector + RRF
Keyword search alone isn't enough. Neither is vector search. What happens when you combine them with Reciprocal Rank Fusion? Hybrid search architecture through the dnomia-knowledge implementation.
Decision Gate: The Missing Piece of Vibe Coding
AI generates code fast, but every accept is a decision. The 8-criteria Decision Gate framework, adapted from Stage-Gate, systematizes technical decisions in AI-assisted development.
Pre-injection vs MCP Tool Loop: Context Strategies for AI Coding Agents
GrapeRoot injects context before the model starts, CodeGraphContext uses MCP tool loops. Same repo, same model, same prompts. One comes out 31% cheaper. Where's the architectural difference? Benchmark data and comparison with my own system.
Which Files Do You Actually Read? Developer Interaction Tracking
Do you know which files you looked at most during a week of work? Which searches returned zero results? Using Claude Code hooks to collect interaction data and trace analytics to reveal patterns, I tracked my own workflow. The data was surprising.
Context Engineering for AI Coding Agents: From Static Documents to a Living Ecosystem
CLAUDE.md and architecture.md are not enough. A four-layer context engineering ecosystem combining semantic code search, knowledge base, decision governance, and learning loops. Based on real project experience.
SUBSCRIBERS ONLY
Local Semantic Code Search: An MCP Server for AI Agents
A local semantic code search MCP server that solves AI coding agents' search problems using Tree-sitter AST chunking and hybrid search. Development process, architectural decisions, and real-world results.
Claude Code Context Management: Three Different Approaches
Claude Code's context window fills up fast. claude-context-mode, claude-context, and dnomia-knowledge solve this problem in different ways. Architectural decisions, trade-offs, and real-world comparison.
AI Pair Programming: Coffee Debt, Gamifying Error Tracking
AI coding assistants make mistakes. Edit commands fail to match, bash scripts crash, users have to step in and correct. The point isn't getting frustrated at errors, it's logging them and extracting patterns. The Coffee Debt system does exactly that.
Why AI Agents Break Files: Practical Strategies and Tests
I benchmarked 5 file editing approaches used by AI coding agents across 20 scenarios. Sequential Edit, Atomic Write, Bottom-up Edit, Script Generation, and Unified Diff, which works when?
RAG Chunking: Strategies, Limitations, and Decision Map
A comprehensive guide to text chunking strategies in RAG systems. 12 different approaches, limitations, multilingual support, and a scenario-based decision table.