skills/compute-mesh/layer-inference-planner/SKILL.md
Plans AirLLM-style layer-by-layer inference across mesh devices for running large models on limited hardware.
npx skillsauth add 47network/Sven layer-inference-plannerInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Plans AirLLM-style layer-by-layer inference for running 70B+ models on constrained hardware. Supports single-device sequential inference and multi-device pipeline distribution with activation transfer estimation.
development
Automated web testing advisor — generate Playwright test scripts, accessibility checks, and performance audits.
testing
Unified security posture reporting — combines SAST, dependency audit, secret scan, infrastructure scan, and pentest results into a single scored report with OWASP and SOC 2 compliance mapping, top risks, and remediation recommendations.
development
Detect hardcoded secrets, API keys, tokens, private keys, and credentials in source code and config files. 20 built-in patterns covering AWS, GitHub, Slack, Stripe, database URLs, JWTs, and more.
development
Static Application Security Testing — scan TypeScript/JavaScript source code for SQL injection, XSS, SSRF, path traversal, command injection, hardcoded secrets, insecure crypto, auth bypass, prototype pollution, and more. 14 built-in rules mapped to OWASP Top 10 and CWE.