Skip to main content
Search

# Agent Skills

31 related articles

Subscribe to updates
  1. NVIDIA releases SkillEvaluator: static quality checks for agent skills, then a with-and-without skill comparison

    NVIDIA has released SkillEvaluator, a tool that checks the quality of AI agent skills and measures their effect on task results. It runs quality checks first, then compares runs with and without a skill; a skill is a folder of instructions and supporting files defined by the Agent Skills specification. The bundled text-analyzer example scores 88.8/100 (B) on the static quality check against a default passing threshold of 70, while publication evidence remains marked INCOMPLETE.

  2. Stitch skills ship as a plugin marketplace: six coding agents can install design and build skills on demand

    The google-labs-code repository has released stitch-skills, a collection of agent skills and plugins for Google Stitch that follows the Agent Skills open standard. It organizes design, build and utility plugins as an addable marketplace for Codex, Antigravity, Gemini CLI, Claude Code and Cursor, while OpenCode requires manually copying skill folders and configuring the Stitch MCP server. The skills depend on a configured and running Stitch MCP server, and selective installs must include the dependencies between skills.

  3. Claude-BugHunter packs bug-hunting know-how into a skill bundle: 83 skills load by topic, no name-calling required

    Sachin Sharma released Claude-BugHunter, a drop-in skill bundle for the Claude Code skills system that splits bug hunting and authorized red-team work into 83 skills and 15 slash commands. The skills are drawn from 681 disclosed reports, 433 of them individually cited and auditable, across 24 core vulnerability classes, with enterprise identity and infrastructure attack matrices. A user describes the target in plain English and the relevant skill loads by topic, with no invocation by name.

  4. NVIDIA open-sources SkillEvaluator: agent skills pass a static quality gate first, then a with-and-without skill run comparison

    NVIDIA released SkillEvaluator, an open-source command-line tool that checks AI agent skills and measures their effect on task results. A skill is a folder of instructions and supporting files defined by the Agent Skills specification. The tool runs a static quality check first, then compares runs with and without a skill; the example skill text-analyzer scores 86.2/100 (B), exits 0, and the default passing threshold is 70. That score covers static quality only, and publication evidence stays marked incomplete.

  5. arXiv

    A few rewritten decision rules in a third-party skill make agents answer wrong while running normally, and most scanners miss it

    Third-party skills in public registries can be exploited by a poisoning method that edits only decision rules while leaving the task and interface intact. Without real user requests or ground-truth answers, StealthyPoison uses synthetic requests and clean executions by a surrogate agent as references to iteratively rewrite formulas, thresholds and branches. Across 566 eligible skills, 5,244 requests and more than 26,000 registry entries, the frozen poisoned skills reached micro-averaged attack success of 31.0%–31.9% on three held-out models, close to the surrogate's 32.9%.

  6. android/skills

    Android skills launches as an official library: development guidance becomes SKILL.md files that agents activate on task

    Google's Android team released a repository called Android skills that turns Android development tasks into modular technical specifications as SKILL.md files following the open agent-skills standard, giving AI agents information on specialized domains and workflows. The skills prioritize use cases where evaluations show large language models underperform, rather than areas where models are already proficient, such as basic Jetpack Compose best practices. Skills install through the Android CLI or the Claude Code and Codex plugin marketplaces, and agents activate relevant ones automatically.

  7. Anthropic

    Agent Skills standard moves to a neutral foundation, with 70-plus platforms and one million skills published

    Anthropic has donated Agent Skills, the open standard it published in December 2025, to the Agentic AI Foundation under the Linux Foundation, transferring the name, logo, GitHub organization and agentskills.io domain. A skill is a folder holding a SKILL.md file that teaches an agent a specific task or domain; the format grew out of Claude Code, which handled general knowledge work but often lacked the procedural knowledge real business tasks need. Since the October 2025 launch, one million skills have been published and installed nearly 280 million times, and more than 70 agent platforms including ChatGPT, Cursor, GitHub Copilot and VS Code have integrated the standard.

  8. anthropics/skills

    Anthropic opens its Claude skills repository: document, PDF and spreadsheet skills go Apache 2.0 as developer references

    Anthropic has published a repository of example skills for the skills system Claude uses: a skill is a folder of instructions, scripts and resources that Claude loads dynamically to complete a specialized task repeatably, such as producing documents under a company's brand guidelines, analyzing data with an organization's workflows, or automating personal tasks. The repository also open-sources, under Apache 2.0, the docx, pdf, pptx and xlsx document skills that power Claude's document capabilities, as a reference for more complex skills in a production AI application; the skills are for demonstration and education only, and behavior received from Claude may differ from what they show, so they should be tested thoroughly before critical use.

  9. expo/skills

    Expo hands AI coding agents an official skills pack: one plugin install, and builds, upgrades and debugging follow Expo's own docs

    The Expo team has released an official skills pack for AI coding agents, giving them focused Expo knowledge: when to use Expo APIs, how to structure common workflows, and which Expo, EAS, React Native, iOS and Android constraints matter. The skills are split into free open-source Expo SDK and React Native skills, skills whose core purpose uses paid Expo Application Services (EAS), and skills for not-yet-finalized Expo APIs, with the same label carried in each skill's description. Expo documentation, Expo CLI and EAS CLI remain the source of truth; the skills help agents apply that guidance correctly.

  10. Cloudflare

    Cloudflare turns agent skills into workspace files: drop a folder, and the agent uses it next turn with no build or deploy

    Cloudflare's Agents SDK now treats Agent Skills as files in the agent's workspace: write a folder containing SKILL.md under /workspace/.agents/skills/ and the agent can use it on its next turn, with no build step or deploy, and the agent can write its own skills. Skills load from a workspace, an R2 bucket, or code, and Think lists every source again before each turn; each SKILL.md is checked against the Agent Skills specification, and a skill that breaks a rule still loads with a logged warning. This is a breaking change: direct agents/skills use, the Vite import, SkillRegistry, and Think's skillWorkspace are removed.