How agent skills work
The mechanics underneath the format: how an agent decides a skill is relevant, what progressive disclosure buys you, and where the security boundaries sit. Useful when something isn't behaving the way you expected.
What Changed in Codex CLI 0.153.0 Through 0.153.2
Codex CLI's 0.153 line added Vim undo and redo, remote plugin marketplace management, a configurable auto-recap setting, and catalog support for GPT-6-Astra.
September 5, 2026
Get Claude Skills
codex cliopenaiclipluginsrelease notes
How Claude Code's /advisor Command Works
Claude Code's advisor tool pairs your main model with a stronger advisor it consults mid-task at decision points, set with /advisor, an advisorModel setting, or a --advisor flag.
September 4, 2026
Get Claude Skills
claude codeadvisormodel configurationcost managementcli
Claude Code's Containment Escape Rule in Auto Mode
Claude Code v2.1.257 added a Containment Escape rule to auto mode's classifier, blocking cloud metadata-credential fetches, egress evasion, and cross-tenant reach unless your environment marks them expected.
September 3, 2026
Get Claude Skills
claude codeauto modesecurityclassifieranthropic
Claude Fable 5.1: What Changed, and What Breaks If You Upgrade
Claude Fable 5.1 shipped 1 September 2026 as the default Fable model, with a quarter-price cache read and three breaking changes for anyone calling Claude Fable 5 today.
September 2, 2026
Get Claude Skills
claude fable 5.1anthropicclaude codeclaude apipricing
Claude's Browser Use Tool vs Computer Use
Claude's browser use tool drives a webpage through its accessibility tree and element references, not just screenshots. Here's how it works and when to reach for it over computer use.
September 1, 2026
Get Claude Skills
claude apibrowser usecomputer useagent toolsanthropic
Why Claude Sonnet 5 Isn't Getting More Expensive on 1 September
Anthropic confirms Claude Sonnet 5's launch pricing of $2/$10 per million tokens is now permanent. The scheduled rise to $3/$15 on 1 September 2026 will not happen.
August 31, 2026
Get Claude Skills
claude sonnet 5anthropicpricingclaude apiclaude code
Agent Skills and the Skills API Are Now Generally Available
Anthropic took Agent Skills and the Skills API out of beta on the Claude API on 19 August 2026. What GA actually changes, what the skills-2025-10-02 beta header still does, and the SDK migration that followed.
August 30, 2026
Get Claude Skills
agent skillsclaude apiskills apianthropicgeneral availability
Cross-Session Messaging Now Works on Bedrock, Vertex, Foundry
Claude Code v2.1.248 brought cross-session messaging to Amazon Bedrock, Google Cloud's Agent Platform, Microsoft Foundry, and sessions with telemetry disabled, providers that were explicitly excluded until now.
August 30, 2026
Get Claude Skills
claude codecross-session messagingamazon bedrockenterpriseanthropic
Claude Code's Live Subagent Streaming to Remote Control
Claude Code v2.1.251 streams a foreground subagent's tool calls and results live to Remote Control clients, while background subagents still show status only. What changed and why it only covers one kind of subagent.
August 30, 2026
Get Claude Skills
claude coderemote controlsubagentsfork modeanthropic
What Claude Code's v2.1.251 Security Fixes Patched
Claude Code v2.1.251 fixed a symlink escape in Read, Write and Edit, a plugin command path-traversal bug, a Workflow tool permission bypass and more. What each fix actually changed.
August 29, 2026
Get Claude Skills
claude codesecuritypermissionsupdatesanthropic
Mid-Conversation Tool Changes on the Claude API
A beta that lets you add or withdraw tools partway through a Claude API conversation without touching the tools array, so the prompt cache stays intact.
August 28, 2026
Get Claude Skills
claude apitool useprompt cachingclaude opus 5agentic loops
Claude Code's SendFeedback Tool: Draft Bug Reports as They Happen
Claude Code v2.1.247 added a SendFeedback tool that drafts a feedback report the moment something goes wrong in a session, for you to review, edit or discard from /feedback.
August 27, 2026
Get Claude Skills
claude codefeedbackdebuggingsettingsanthropic
Claude Code's /web-setup and /status: Checking Your GitHub Connection
Claude Code v2.1.243 added a /status line showing whether GitHub is connected for Claude Code on the web, plus a tip pointing you to /web-setup. Here is what each connection method actually does.
August 27, 2026
Get Claude Skills
claude codeclaude code on the webgithubcloud sessionsanthropic
The Advisor Tool on the Claude Messages API
How the Claude Messages API's advisor tool pairs a cheap executor model with a stronger advisor for mid-generation guidance: setup, result variants, cost and model compatibility.
August 26, 2026
Get Claude Skills
claude apimessages apianthropicagent skillscost management
How Budgets Work in Claude Managed Agents
How Claude Managed Agents session and deployment budgets cap spend at public list rates: max_list_cost, the budget_reached stop reason, and multiagent sharing.
August 26, 2026
Get Claude Skills
claude apimanaged agentscost managementanthropicbudgets
Multiagent Orchestration in Claude Managed Agents
How a Claude Managed Agents coordinator delegates to a roster of other agents: session threads, the advisor pattern, MCP server scoping, and the hard limits on delegation.
August 25, 2026
Get Claude Skills
claude apimanaged agentsmultiagentanthropicagent skills
Claude Managed Agents Explained
Claude Managed Agents is Anthropic's hosted agent harness: versioned agent configs, sandboxed sessions, and skills loaded from a GitHub repo. Here is how it differs from the Messages API.
August 24, 2026
Get Claude Skills
claude apimanaged agentsagent skillsanthropicsandboxes
Codex CLI 0.148.0: Session Branching and Export
Codex CLI 0.148.0 added codex exec fork for headless session branching, /export to Markdown, session archiving, thread cost visibility, Bedrock as a provider and async hooks.
August 24, 2026
Get Claude Skills
codex cliopenaichangelogsessionsdeveloper tools
Anthropic's Open-Source Claude API Skill Explained
The claude-api skill ships bundled with Claude Code and installs anywhere Agent Skills work, giving Claude current reference material for the Messages API and Claude Managed Agents.
August 23, 2026
Get Claude Skills
claude apiagent skillsclaude codeanthropicmanaged agents
Claude Code's Data Residency Pricing Premium Explained
Claude Code v2.1.239 made /cost, the status line, and --max-budget-usd account for the 1.1x US-only-inference premium on data-residency workspaces. Here's what that premium is.
August 23, 2026
Get Claude Skills
claude codepricingdata residencyanthropiccost management
Codex CLI's Interactive Task Dashboard: What's New in 0.149.0
Codex CLI 0.149.0 added a codex agents dashboard for managing tasks, plus /cd, /pwd and /cwd, codex queue, expanded Vim editing and a more capable codex doctor. Here's what actually changed.
August 23, 2026
Get Claude Skills
codex cliopenaichangelogcli toolsdeveloper tools
Claude Code Now Resumes Automatically When Your Usage Limit Resets
Claude Code v2.1.234 added automatic continuation when a session or usage limit resets, turned on by default and controlled from /config, plus a separate Desktop app checkbox with its own mechanics.
August 22, 2026
Get Claude Skills
claude codeusage limitsclaude code desktopsettingsanthropic
Get Notified When Another Claude Code Session Goes Idle
notify_when_idle lets one Claude Code session ask another on the same machine to send one notice the moment it finishes, replacing manual polling with a one-shot, opt-in subscription.
August 22, 2026
Get Claude Skills
claude codecross-session messagingsendmessagebackground tasksanthropic
Claude's Beta Skill and Plugin Security Scanning, Explained
Claude Enterprise organisations can now turn on automated scanning that checks custom skills for malicious behaviour before members can use them. Here is exactly what it covers and what it doesn't.
August 21, 2026
Get Claude Skills
agent skillssecurityclaude enterpriseclaude coworkplugins
Explainer
How MCP's Tasks Extension Works
MCP's Tasks extension lets a server hand back a task ID instead of blocking on a slow tool call, so a client can poll for the result later. Here is the actual mechanics: states, methods and the polling flow.
August 21, 2026
Get Claude Skills
mcpmodel context protocolagent skillsasync tasksprotocol spec
Why Claude Code Shortens or Drops Your Skill's Description
Claude Code's skill listing has a budget scaled to about 1% of context. Once it overflows, your least-invoked skills lose their descriptions first, a common reason a skill quietly stops triggering.
August 20, 2026
Get Claude Skills
claude codeagent skillsskill.mdcontext windowskill authoring
Explainer
What Changed in the MCP 2026-07-28 Spec
The Model Context Protocol's 2026-07-28 revision drops the stateful handshake for a stateless core, adds mid-call user input, and deprecates Roots, Sampling, Logging and the old HTTP+SSE transport.
August 20, 2026
Get Claude Skills
mcpmodel context protocolagent skillsprotocol spec
Claude Code Channels: Pushing Events Into a Session
Channels push messages, CI results and webhooks straight into a running Claude Code session so it can react while you're away, instead of Claude polling for updates on a timer.
August 19, 2026
Get Claude Skills
claude codechannelsmcpautomationci
What Claude Code's Monitor Tool Does
Monitor lets Claude Code watch a log, a CI job, a directory or a WebSocket feed in the background and react the moment something changes, instead of you re-prompting on a timer.
August 19, 2026
Get Claude Skills
claude codemonitor toolbackground tasksautomationwebsocket
How Claude Code's Subagent Forking Works
Subagent forking lets a subagent inherit the full parent conversation and prompt cache instead of starting fresh, and it now runs in the background by default. Here's exactly how it works.
August 19, 2026
Get Claude Skills
claude codesubagentsagent skillsbackground tasksanthropic
What Claude Code's Agent Teams Feature Does
Agent teams let several Claude Code sessions coordinate through a shared task list and direct messages, with one acting as lead. Here's how it actually works, and where it's still experimental.
August 18, 2026
Get Claude Skills
claude codeagent teamssubagentsmulti-agentanthropic
How Claude Code's /goal Command Works
/goal sets a completion condition and keeps Claude working across turns until a separate evaluator model confirms it's met. Here's exactly how the evaluation loop works, and when to reach for it instead of /loop.
August 18, 2026
Get Claude Skills
claude codegoal commandautomationauto modeanthropic
Claude Code's GitLab Support: What Changed
In two releases on 13 and 14 August 2026, Claude Code added GitLab merge request support to --worktree and agent view, secret redaction for GitLab tokens, and GitLab plugin marketplaces.
August 17, 2026
Get Claude Skills
claude codegitlabworktreespluginsanthropic
GitHub Copilot Pricing: Which Tier Do You Actually Need?
GitHub Copilot has four individual tiers as of August 2026, Free, Pro, Pro+ and Max. Here's exactly what each includes, and which one you need for agent skills and third-party coding agents.
August 17, 2026
Get Claude Skills
github copilotpricingagent skillsagent modethird-party agents
Codex CLI Can Now Import Your Cursor Skills
OpenAI's Codex CLI 0.147.0 extended its /import command to pull in Cursor-managed skills, not just Claude Code's. Here's what actually moves, and what to check before you trust it.
August 15, 2026
Get Claude Skills
codex clicursoragent skillsskill.mdportability
How Claude Code Sessions Message Each Other
Claude Code sessions can now message each other with ListAgents and SendMessage. How discovery, delivery, and inbound controls work, and how it differs from subagents.
August 14, 2026
Get Claude Skills
claude codecross-session messagingsubagentsremote controlanthropic
Claude Code's Auto Mode Is Now the Default: What Changed
From 14 August 2026, auto mode replaces manual permission prompts by default for new Claude Code sessions on Pro, Max and Team. Here's what the classifier actually allows and blocks.
August 13, 2026
Get Claude Skills
claude codeauto modepermissionsanthropic
Agent Plugins Explained: The New Skills + MCP Standard
A new open standard from OpenAI, Amazon, Microsoft, Cursor and Vercel packages Agent Skills and MCP servers into one distributable plugin. Here's what it actually changes.
August 12, 2026
Get Claude Skills
agent pluginsskill.mdmcpagent skillsopen standard
Agent Skills in 2026: The State of the Ecosystem
The standard shipped in late 2025. 2026 is when it spread, Cowork, Record a Skill, a cross-agent installer, and tens of thousands of public skills. Here's where things actually stand, opinions included.
August 10, 2026
Get Claude Skills
agent skillsecosystemclaude coworkskills cliai agents
Claude Opus 5 and Agent Skills: What Changed
Opus 5 shipped 24 July 2026 with real gains in agentic coding and multi-agent coordination. Here's what that actually means for how you write and structure skills.
July 27, 2026
Get Claude Skills
claude opus 5agent skillsclaude codemulti-agentclaude managed agents
GPT-5.6, Codex and Agent Skills: Where OpenAI Stands
OpenAI's own models now read the same SKILL.md format Anthropic published. Here's the 2026 model timeline, what a 1M-token context window actually changes for skills, and how Codex CLI handles them.
July 1, 2026
Get Claude Skills
gpt-5.6codex cliopenaiagent skillscontext window
Agent Skills Security: What to Check Before You Install
What to check before installing an agent skill: bundled scripts, permissions, prompt injection risk, and a concrete pre-install checklist.
April 14, 2026
Get Claude Skills
securityagent skillsprompt injectionclaude codesupply chain
Explainer
How AI Agents Discover and Activate Skills
Discovery, activation, execution: the three-phase pipeline behind every agent skill, and why the description field is the entire discovery surface.
February 24, 2026
Get Claude Skills
agent skillsprogressive disclosureskill activationskill.mdexplainer
The SKILL.md Format Explained
A field-by-field breakdown of the SKILL.md format: frontmatter, body structure, and the scripts, references and assets folders a skill can bundle.
January 27, 2026
Get Claude Skills
skill.mdagent skillsyaml frontmatterfile formatclaude code
What Are Agent Skills? The Complete Guide
Agent Skills are portable folders of instructions that let AI agents like Claude Code and Cursor load expertise on demand. Here is how the format works, end to end.
January 14, 2026
Get Claude Skills
agent skillsskill.mdclaude codeai agentsprogressive disclosure