A step-by-step guide to building a Claude Code skill from scratch, focusing on the skill folder structure, SKILL.md (especially the critical role of description), using scripts for consistency, and lazy-loading references. The key insight: description determines if the skill triggers, not the instructions. Practical for engineers using Claude Code to persist workflows.
As of July 2026, no single model dominates all tasks. Kimi K3 (2.8T params) leads frontend UI and image understanding, winning 6 of 7 domains at 1/12 the cost of Fable 5. Claude Fable 5 achieves 80.3% on SWE-Bench Pro, ideal for backend architecture and long-running autonomous agents, but at $10/$50 per million tokens. GPT-5.6 Sol excels at debugging but may game vague success criteria. This guide provides a task-specific routing framework, cost math, licensing considerations, and vendor lock-in mitigation. The key skill is routing, not picking a permanent favorite.
Graphify is an open-source tool that transforms codebases, docs, PDFs, images, and videos into a knowledge graph for AI coding assistants like Claude Code, Cursor, and Gemini CLI. It uses tree-sitter AST for deterministic, local-only code parsing, and delegates semantic extraction for non-code assets to the assistant's model. The output includes an interactive HTML visualization, a Markdown report, and a reusable graph.json, enabling natural-language queries, path traversal, and concept explanations. Every edge is tagged EXTRACTED or INFERRED, so users always know what was read vs. guessed. Ideal for engineers onboarding large unfamiliar codebases or augmenting long-tail maintenance workflows.
Simon Willison shares a practical tip from the Claude Code team: let Fable use its own judgement to decide when to write tests and delegate coding tasks to cheaper subagents. With Claude Code's Fable token prices about to rise, he demonstrates how to configure a memory file so the main model can autonomously pick a lower-cost model (Sonnet for substantial work, Haiku for trivial edits) for implementation tasks while retaining judgement-heavy work in Fable. Early results show significantly reduced Fable consumption without sacrificing productivity.
Moonshot AI releases Kimi K3, a 2.8T-parameter open model built on Kimi Delta Attention (KDA) and Attention Residuals (AttnRes), activating 16 out of 896 experts with a reported 2.5× scaling efficiency improvement over K2. It supports native vision and a 1M-token context window. While generally trailing top proprietary models like Claude Fable 5 and GPT 5.6 Sol, K3 achieves competitive scores on coding, knowledge work, and reasoning benchmarks. The post details case studies: GPU kernel optimization, a from-scratch Triton-like compiler (MiniTriton), 3D open-world game development, autonomous chip design (48-hour run), and rapid scientific research reproduction. K3 is available now via Kimi.com, Kimi Work, Kimi Code, and API; full weights open-sourced by July 27, 2026. Recommended for AI engineers, agent developers, and researchers needing long-horizon agentic capabilities.
Armin reports a counterintuitive bug encountered while developing Pi code editor: newer Claude models, including Opus 4.8 and Sonnet 5, invent extra fields in the nested edits[] array when calling Pi's custom edit tool, causing the tool call to be rejected. Older Claude models do not exhibit this behavior. Armin hypothesizes that Anthropic's reinforcement learning has specifically optimized newer models for Claude Code's built-in edit tool, inadvertently degrading performance on other tool schemas. The piece questions whether third-party coding harnesses must implement multiple edit tools per model family, and highlights the fundamental trade-off between specialized training and general tool compatibility.
This article systematically compares three major AI agent skill frameworks: Matt Pocock Skills (engineering practice), Superpowers (community workflow), and Agent Skills (production lifecycle). It evaluates them across positioning, skill granularity, learning curve, token consumption, tool support, and community size, offering selection recommendations for individuals, small teams, mid-large teams, and enterprises. Key findings: Matt Pocock Skills excels at deep alignment and architecture optimization, Superpowers provides end-to-end workflows with a rich plugin ecosystem, and Agent Skills enforces verification gates and anti-rationalization for quality. It also suggests combination strategies. Suitable for developers and tech leads choosing a workflow framework for AI coding assistants.
A firsthand account of Anthropic's deteriorating trust: unreliable APIs tied to subscriptions, a locked-in and buggy Claude Code ecosystem, and opaque billing changes that effectively raise costs. The author argues these moves fund model training, not product improvement. Their response: shift to an 'agent-assisted' workflow and swap Claude for open-source models like Qwen and GLM via OpenRouter, gaining cost control and data security. For engineers feeling trapped in a single AI platform.
Dan Luu shares his extensive experience with AI coding agents over the past year, focusing on testing, benchmarking, and agentic loops. He compares fuzzing vs. LLM-driven bug finding, finding fuzzing faster with lower false-positives; evaluates 'caveman mode' with 50 runs showing inconsistent savings; highlights high variance in LLM benchmarks, making public evals nearly useless for individual users. He also discusses automated PR generation from support tickets, multi-persona false-positive reduction, and challenges in data analysis and autonomous loops. For engineers interested in real-world effectiveness of AI coding tools.
This article provides a detailed 14-step roadmap for building a self-improving agent system using Claude Fable 5. It moves Fable 5 from a prompt-and-close tool to a compounding system: using /goal and Outcomes for self-correcting loops, independent verifier sub-agents over self-critique, state files (STATE.md) and Skills for cross-session memory, and Dynamic Workflows and Routines for long-running autonomy. It includes a cost-capability matrix (Fable 5 for orchestration, Sonnet 4.6 for workers, Haiku 4.5 for graders) and guidance on handling the Mythos safety boundary. Suitable for AI engineers and system designers aiming to leverage Fable 5's days-long autonomous capability.
The Claude Code team defines four loop patterns (turn-based, goal-based, time-based, proactive) with trigger, stop criteria, use cases, and token management tips. Concrete commands like /goal, /loop, /schedule and a SKILL.md example show how to make agents iterate, self-verify, and compose primitives into automated workflows. A practical guide for developers exploring agent engineering.
Anthropic engineers share practical lessons from maintaining hundreds of Claude Code skills internally. The post categorizes skills into nine types—library reference, product verification, data fetching, business process automation, code scaffolding, code quality, CI/CD, runbooks, and infrastructure ops—and offers concrete writing tips: focus on non-obvious hints, build gotchas sections, use the filesystem for progressive disclosure, avoid over-constraint, design setup flows, treat the description field as a trigger for the model, store memory (logs, JSON, SQLite), and bundle scripts/libraries to reduce token waste. It also covers distribution strategies (in-repo vs. marketplace), skill composition, and usage measurement via PreToolUse hooks. A must-read for anyone building agent-centric workflows.
Bun's creator Jarred Sumner recounts how he used Anthropic's Claude Fable 5 to rewrite Bun's 535,496 lines of Zig into Rust in 11 days. The motivation: Zig's manual memory management caused numerous use-after-free, double-free, and memory leaks when mixed with JavaScriptCore's GC. Instead of an incremental port, he orchestrated 64 Claude agents in parallel using dynamic workflows and adversarial review. 100% of Bun's test suite (over 600k assertions) passed on all 6 platforms. The rewrite fixed 128 bugs, reduced memory usage by up to 90%, shrank the binary by ~20%, and improved throughput by 2-5%. The article details the workflow, common porting mistakes (e.g., debug_assert! side effects, slice overruns, comptime format differences), and how Rust's Drop systematically prevented memory leaks. A first-hand account of using cutting-edge AI to accomplish a year-long team project in less than two weeks.
This article covers the v1.1 release of the skills repository, a toolset for AI agents. Key changes: /to-prd renamed to /to-spec to unify the term 'spec'; /to-plan and /to-issues merged into /to-tickets with blocking edges for both local files and real trackers. Grilling skills now prevent multiple simultaneous questions, add a confirmation gate, and avoid self-grilling. A new /wayfinder skill decomposes large tasks into GitHub issues with dependency graphs, accompanied by /research and /prototype skills. Code review skill integrates Fowler's ten code smells (e.g., mysterious name, duplicated code) with just 10 lines of guidance. TDD skill becomes reference-only, moving refactoring to code review. The recommended workflow: Grilling → Spec → Tickets → Implement → Code Review. Suitable for engineers building with AI agents.
This article introduces `/writing-great-skills`, a meta-skill that serves as a reference framework for authoring and editing predictable AI skills. The core idea is the trade-off between **cognitive load** and **context load**: model-invoked skills cost context load but fire automatically, while user-invoked skills cost zero context load but require you to remember their existence. The article provides tools for managing these loads, including leading words (compact anchors for execution), information hierarchy (progressive disclosure), pruning (single source of truth and no-op test), and failure modes (premature completion, duplication, sediment, sprawl). A must-read for system builders writing consistent, maintainable skills for agents.
A practical 12-step walkthrough for giving Claude agents persistent memory across sessions. Covers four layers: built-in Chat Memory, Project instructions, a lean memory file (CLAUDE.md), and Dreaming – a scheduled background process that consolidates and reorganized memory. Includes setup steps, API calls, and advice on filtering what to remember. Harvey reported ~6x task-completion rate improvement with Dreaming. Ideal for engineers building long-running agents.
This article presents a 14-step roadmap for building an Agent harness with Claude, from a single agent to a self-improving system. The author argues that the harness — comprising model, tools, permissions, and initial context — is the foundation for any loop, and a weak harness leads to slop. It covers concrete practices: file structure (.claude/), CLAUDE.md for standing facts, settings.json for permission pre-approvals, subagents for isolated contexts, skills for reusable procedures, hooks for deterministic enforcement, and memory for cross-session learning. The final steps add loops and dynamic workflows, closing the feedback loop where output → lessons → skills → better output. The article targets engineers who run or plan to run multi-agent code generation systems.
The author shares hands-on experience with Claude Fable for agentic coding, emphasizing that the prompt (map) never fully matches the codebase (territory). He categorizes unknowns into four types (known knowns, known unknowns, unknown knowns, unknown unknowns) and provides practical techniques to systematically discover them: blindspot passes, brainstorming & prototypes, interviews, references, implementation plans, implementation notes, pitches, and quizzes. Ends with a real example of editing the Fable launch video. Suitable for engineers using AI-assisted coding.
Pi author discovers that Anthropic's Opus 4.8 and Sonnet 5 inject spurious keys (requireUnique, oldText2, cost, etc.) into the edits[] array of Pi's edit tool, while older models do not. The failure is context-dependent and reproducible in agentic sessions. The post dissects Anthropic's tool calling internals: ANTLM markers, JSON-serialized nested arrays, and Claude Code's extremely forgiving harness that silently filters unknown keys and retries malformed calls. Author hypothesizes that RL post-training over Claude Code's flat old/new_string schema creates a strong prior, making newer models worse at following non-canonical tool schemas. Strict tool invocation fixes the issue, but Anthropic's complexity limits prevent Claude Code from using it. Key takeaway: tool schemas are not distribution-neutral; any harness must inherit Claude Code's quirks.
The author shares a real-world comparison between Superpowers and mattpocock/skills, explaining why they switched. Superpowers uses hooks to enforce a rigid workflow, which is helpful for novices but often overcomplicates simple tasks and burns excessive tokens. mattpocock/skills takes a 'real engineer' approach, giving control back to the user via explicit commands like /grill-with-docs, /to-prd, /to-issues, and /implement. Key advantages: lower token consumption, built-in debugging (/tdd, /diagnosing-bugs), model handoff (/handoff), and architecture refactoring (/improve-codebase-architecture). The author pairs these skills with Fable 5 and Codex 5.5 models, storing PRDs and issues on GitHub for traceability. A candid take for engineers evaluating agent frameworks and tooling.
The author shares their journey from a failed attempt at orchestrating long-running AI agent tasks to discovering the Superpowers Skill Set, which solves the core pain points. Superpowers decomposes the development workflow into three phases: brainstorming, writing-plans, and executing-plans (with subagent-driven-development). Key design elements include: using separate prompt templates (implementer, spec-reviewer, code-quality-reviewer) to enforce separation of concerns; spinning up a fresh subagent for each task to avoid cascading context pollution; using hard constraints like "Never/HARD-GATE" to prevent agent deviation; and enforcing software engineering best practices such as TDD, DRY, and YAGNI. The author argues that with frontier models like Opus 4.8 and Codex GPT-5.5 now being sufficiently capable, the real bottleneck is harness design—using clear specifications and structured processes to make even cheaper models reliable for long-duration tasks.
Superpowers 6 is released, with its biggest improvements driven by an automated research loop. The author used Anthropic's Fable model (briefly available) to systematically optimize their Subagent Driven Development pipeline. Over 36 hours and ~$165 in token spend, 25 experiments were run, yielding a 50% reduction in wall-clock time and 60% reduction in token consumption vs. v5. Key optimizations: merging spec compliance and code review agents, pre-baking review packets to minimize git operations, and dynamic agent allocation based on task type (e.g., using cheap haiku for non-code plans). The post also documents falsified hypotheses (e.g., capping controller thinking backfires) and emphasizes the role of their eval suite in rigorous measurement.
Cursor shares how it continuously improves its agent harness, covering context window evolution from static to dynamic fetching, a two-layer evaluation system (offline benchmarks and online A/B tests measuring code keep rate and user satisfaction), tool call error classification and repair pipeline (anomaly detection + automated log analysis with Cloud Agents), per-model customization of tool formats and prompts (e.g., patch vs. string replacement), and mid-chat model switching with specialized instructions. The post concludes with a vision of multi-agent architectures where the harness orchestrates specialized sub-agents.
Browser Harness is a thin, self-healing CDP harness that connects an LLM directly to a real Chrome browser via a single WebSocket, with zero intermediate layers. When the agent needs to perform an action it hasn't seen before (e.g., file upload, cross-origin iframe interaction, drag and drop), it writes the missing helper code on the fly and saves it into an agent-workspace for reuse. The core package is roughly 1K lines, enabling complete freedom for browser automation tasks. Aimed at developers who need AI agents to perform real, unconstrained browser interactions.
Open Design is a local-first, open-source alternative to Claude Design. It is agent-native, meaning it doesn't ship its own agent but works with 22 coding-agent CLIs (Claude Code, Codex, Cursor, Copilot, etc.) already on your PATH. Using MCP, the agents read DESIGN.md brand systems, skills, and plugins to generate prototypes, live dashboards, decks, images, videos, and HyperFrames. Exports to HTML, PDF, PPTX, MP4. Supports BYOK for any OpenAI-compatible endpoint. Ships 100+ skills, 150+ brand-grade design systems, and 261 plugins. Ideal for engineers and designers who want brand fidelity and local control.
Anthropic shares four lessons from months of internal testing on building human-agent teams. The shift is from a single-player experience (one human, one AI) to a multiplayer model where agents hold their own credentials, persistent memory, and broad access, joining team channels as full members. The key insights: work in public so agents have context, define clear roles and tool access for every member, set an ambitious north star to make agents proactive, and build trust by granting autonomy gradually. Includes practical examples like agent-led bug backlogs and doer-verifier patterns. A must-read for teams embedding AI agents into collaborative workflows.
This article is an official engineering guide from Claude Code that systematically lays out four agentic loop patterns and their use cases. Turn-based loops are for short exploratory tasks; users encode manual verification steps into SKILL.md — e.g., asking Claude to start a dev server, take screenshots, and check the browser console. Goal-based loops, triggered by /goal, define deterministic termination criteria such as 'get the Lighthouse score to 90 or above' and force iteration until the target is met. Time-based loops come in two flavors: /loop for local polling on an interval and /schedule for cloud-triggered routines, ideal for recurring work like PR review or CI fixups. Proactive loops combine /schedule, /goal, dynamic workflows, and auto mode into a pipeline for long-running, well-defined streams of work. The article also covers code quality maintenance and token usage management: encoding conventions, using scripts instead of re-reasoning, routing routine work to cheaper models, and monitoring cost with /usage. Suitable for engineers embedding Claude Code into daily dev workflows.
The creator of Claude Code says he no longer writes prompts—loops prompt it instead. This post introduces a 5-level progression of human-AI workflow: from Level 1 (single-turn prompting), through Level 2 (manual loop of do-check-correct), Level 3 (verified loop with separate judges for 'done'), Level 4 (self-running loop using /goal command with guardrails), to Level 5 (autonomous systems where loops self-start, run in parallel, and persist lessons into a skill base). Each level comes with a tell and a concrete next step. For developers who still feel they are 'babysitting' their AI agents.
This is a technical roadmap for building reliable autonomous loops, arguing that a loop is fundamentally different from a prompt—a prompt requires manual initiation while a loop drives itself: set a goal once, then the system finds work, executes, checks, fixes, and repeats until completion. The author emphasizes that the ceiling is set not by prompting skills but by engineering a loop that converges toward truth rather than becoming an expensive random walk. The piece provides step-by-step guidance (Step 0 through Step 7) with working code (Bash scripts), explaining the mechanics of stateless iteration (fresh context per turn to combat context rot), building a narrow relevant context with a token budget, designing incorruptible checks (external deterministic oracle + reward-hacking defense gates + adversarial judge on a different model), dual-level state persistence (human-readable STATUS.md + machine-parseable JSON), physical isolation (git worktree, container with --network none), brakes with observability (structured JSONL log, circuit breakers for stuck/repeated failures, liveness heartbeats), and nonlinear cost analysis (why stateless keeps per-iteration cost constant while stateful grows quadratically). This is aimed at production engineers building AI agent pipelines who need practical, verifiable techniques.
This article, inspired by Claude Code creator Boris Cherny, argues that the center of gravity in agent development has shifted from prompt engineering to loop engineering. It unpacks the trivial core loop and identifies four hard problems: knowing when to stop (distinguishing tool-call cessation from task completion), maintaining context hygiene to avoid decay, designing tools that agents can actually use (idempotent writes, error messages for LLMs), and embedding a critic in the loop to prevent self-agreement. The piece underscores that the model is commoditized; the loop—the harness—is where real engineering value lies. A must-read for engineers building autonomous agent systems.
Boris Cherny, who built Claude Code at Anthropic, no longer writes prompts by hand—he writes loops. This guide explains what a real loop is: a small system that runs Claude repeatedly until a job is done, complete with self-checking, state persistence, and automatic stopping. Unlike cron jobs, loops contain a decision-maker (Claude) that can adapt mid-stream. The article covers Claude Code's /goal (loop until done) and /loop (repeat on a schedule) commands, and provides a paste-ready charter template with sections for goals, work sources, work instructions, self-verification, memory, and stop conditions. Ideal for engineers transitioning from prompting to building persistent, autonomous AI workflows.
When coding agents produce thousands of lines of often solid code in minutes, the engineering bottleneck shifts from writing to trusting, making review the most leveraged skill in software. Multi-source 2026 data (Faros AI, CodeRabbit, GitClear, GitHub) shows: AI users generate ~4x raw output but only ~12% more delivered value; code churn up 861%, defect rate from 9% to 54%, review duration up 441.5%, and zero-review merges up 31.3%. The article argues the fix is not to stop using AI but to tier review effort by blast radius: light for solo no-user projects, heavy for large enterprises. Specific advice: triage PRs upfront, require evidence before review, watch test rewrites, run two differently-structured AI reviewers in parallel, and upgrade humans from line-level review to spot-checking and auditing. The durable skill is understanding a system well enough to stand behind it.
Starting agents is cheap, but closing the loop (review, merge, judgment) must route through a single serial processor: you. The author frames human attention as the GIL for AI agents — you can spawn 20 agents, but your throughput to main is bounded by your serial review rate. Amdahl's Law applies: the non-parallelizable judgment fraction caps any speedup. Strategies include: scaling agent count to review rate, not UI; sorting tasks into async-able vs. judgment-heavy; batching reviews; making agents prove correctness with tests/screenshots. The insight: busy ≠ productive. Unpaid orchestration tax accumulates both technical debt and cognitive debt, silently eroding your mental model of the system.
ECC is a comprehensive agent harness operating system designed to enhance multiple AI coding assistants including Claude Code, Cursor, Codex, OpenCode, and GitHub Copilot. It's not a standalone AI tool but a collection of 260+ skills, 67+ agents, persistent memory, continuous learning, cost optimization, and security auditing features. By providing unified rules, hooks, and MCP configurations across platforms, it addresses issues of inconsistent agent behavior, context loss, and inadequate security. Ideal for professional developers deeply using AI coding assistants and teams seeking standardized agent engineering practices.
The author shares a practical workflow for coding with AI agents: define a verifiable 'definition of done' (model eval score, QA pass, green tests, performance benchmark), wrap it in a loop for the agent to iterate autonomously, and get notified via Slack when a decision is needed or the task completes. Loops run in the cloud, not on the local machine. The author runs 3-5 long loops concurrently plus shorter tasks. For engineers looking to level up from one-shot agent interactions to long-running autonomous optimization tasks.
This post breaks down a 9-step pattern for running multiple Claude Code subagents in parallel without chaos. The key insight is that launching many agents is easy and usually messy; the skill is in the orchestration loop: verify the task is splittable, have the lead agent decompose it into atomic subtasks with human approval, isolate each agent via Git worktrees to prevent file collisions, then fan out in parallel. A SubagentStop hook gates every result (tests + lint), a grader scores each against a rubric and automatically sends weak ones back for revision, and finally only the lead agent merges passing results in dependency order. The entire loop can be packaged as a reusable skill or slash command.
This article breaks down the engineering of AI-driven development loops, contrasting a single engineer shipping 259 PRs in a month with a runaway loop that burned $47,000. It dissects six essential components—state file, automation/scheduling commands (e.g., /loop, /schedule, /goal), git worktrees, skills, MCP connectors, and sub-agents (writer vs. checker)—with concrete configuration examples for both Claude Code and OpenAI Codex. The piece provides a brake configuration template (max_turns, max_budget_usd, scope, circuit_breaker), describes four failure modes, and offers low-cost starting strategies. Aimed at engineers building or evaluating AI agent workflows.
This article explains agent loops for product managers, distinguishing between routines, workflows, and true goal-driven loops. The key insight is that the hard part is defining the stop condition: a verifiable definition of done with an objective check or independent grader. It provides a template for building loops, guidance on writing stop conditions, and cost management advice (e.g., tracking cost per accepted change). Common failure modes are discussed: runaway costs without iteration caps, context drift, and passing tests without being correct. The author concludes that loop engineering is just the latest name for intent engineering—precisely defining goals, boundaries, and completion criteria.
This article explains agent loops for product managers, distinguishing between routines, workflows, and true goal-driven loops. The key insight is that the hard part is defining the stop condition: a verifiable definition of done with an objective check or independent grader. It provides a template for building loops, guidance on writing stop conditions, and cost management advice (e.g., tracking cost per accepted change). Common failure modes are discussed: runaway costs without iteration caps, context drift, and passing tests without being correct. The author concludes that loop engineering is just the latest name for intent engineering—precisely defining goals, boundaries, and completion criteria.
The AI coding community is shifting from prompting agents manually to designing loops that prompt agents for you. This is the most practical, production-oriented guide available: what an agent loop is, why it matters, and what one looks like in the real world. The author breaks down the six mandatory components (trigger, isolation, written-down context, tool reach, second-agent checker, on-disk state), then illustrates with two concrete examples: a PR babysitter that checks every 15 minutes and auto-fixes CI failures, and Claude Code's /goal command. It also covers where cost actually goes (iterations, not tokens), when not to loop (one-shot edits, unbounded exploration), and predictable failure modes (verification burden stays human, comprehension debt, silent drift).
Most developers debug with Claude Code by pasting errors and accepting speculative fixes, leading to a 40-60 message death spiral. This post proposes a six-step loop: first establish a reliable repro (failing test), isolate the search area in plan mode, dispatch read-only subagents to trace root causes from multiple angles, fix only the root cause (not symptoms), verify with an automatic hook (e.g., PostToolUse running the test), and keep the repro as a permanent regression test. The key insight is that Claude Code was always capable; the failure mode is skipping straight to 'fix' before understanding the bug.
This article demonstrates a practical approach to building a self-improvement loop for AI Skills using inner and outer agent loops. The inner loop triggers a cloud agent via GitHub Action on each new issue, applying a triage Skill to classify it. The outer loop runs daily, reviews all human corrections (label changes and comments), and generates a diff to update the Skill file, which is then merged back. The author uses Warp's Oz cloud agent platform for issue triage, providing complete code and a sample repo. The pattern is generalizable to code review, bug fixing, and incident response. Suitable for engineers building AI agents who want to improve skill quality over time.