
Product
Socket Threat Intel Now Shows Which Malware Campaigns Were Blocked or Detected
Socket Threat Intel now shows which malware campaigns were blocked or detected in your organization, with links to scans and Firewall events.
@aicf/agent-skills
Advanced tools
Agent Skills for building, integrating, testing, auditing, and releasing AICF-based AI capability systems.
Reusable Agent Skills for coding agents that build, integrate, test, audit, and release AI Capability Framework projects.
These are builder skills. They guide Codex or another compatible coding agent while it edits a repository. They are not AICF runtime capabilities, provider tools, application-agent tools, or a replacement for AICF validation.
The package contains 17 public skills:
See docs/skill-index.md for descriptions and trigger examples generated from the current package.
Run package commands from this directory:
cd agent-skills
npm install
npm run check
npm run pack:dry
The package is independent from the root AICF package. The repository root does not use npm workspaces for this package.
npm run validate
npm run check:triggers
npm run index
npm run check:index
npm run test
npm run check:public
npm run check:release
npm run check
npm run pack:dry
node scripts/aicf-skills.mjs list
Expected successful output includes messages such as Skill validation passed,
Trigger coverage passed, Skill index is current, and
Agent skills release readiness passed.
The root repository exposes equivalent script hooks such as npm run skills:ci,
npm run skills:check, npm run skills:pack:dry, and npm run skills:publish:dry.
Copy the current skills to a local discovery folder:
node scripts/install-skills.mjs --target ../.agents/skills
node scripts/install-skills.mjs --target $HOME/.agents/skills
The install script copies only skill folders. It does not copy package metadata, tests,
plugin metadata, dependencies, or local artifacts. It refuses overwrite unless --force
is passed.
The Codex plugin manifest lives at .codex-plugin/plugin.json and points to ./skills/
with relative asset paths. The package is released as @aicf/agent-skills and uses the
same version and release tag as the root ai-capability-framework package.
Prerelease install example:
npm install @aicf/agent-skills@next
npx aicf-skills list
Before the first real scoped-package publish, the maintainer must configure npm
ownership and Trusted Publishing for @aicf/agent-skills.
Reference docs:
New or changed skills must keep the required frontmatter, section order, one-level references/assets, trigger fixtures, and public-safety rules. Run:
npm run index
npm run check
All skill content must be public-safe. Do not add private specs, local paths, prompt
text captured from a real session, provider payloads, traces, credentials, account IDs,
tenant IDs, or customer data. Use synthetic examples and example.com where examples
need URLs.
MIT, matching the root AI Capability Framework v1 license decision.
FAQs
Agent Skills for building, integrating, testing, auditing, and releasing AICF-based AI capability systems.
We found that @aicf/agent-skills demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 1 open source maintainer collaborating on the project.

Product
Socket Threat Intel now shows which malware campaigns were blocked or detected in your organization, with links to scans and Firewall events.

Security News
Fixes for eMudhra and Brotli are out. Getting them everywhere they’re needed means tracking down bundled libraries, runtimes, and certificate stores.

Threat Research
/Security News
Socket uncovered secret-stealing workflows that ran in LiteLLM’s GitHub repository 11 days before the supply chain attack on PyPI and Docker.