
erichowens/some_claude_skills
2 skills796 installs231 starsGitHub
Install
npx skills add https://github.com/erichowens/some_claude_skillsSkills in this repo
1Video Processing Editingvideo-processing-editing is an agent skill from erichowens/some_claude_skills that helps solo builders handle repetitive video processing and editing tasks through procedural guidance in the agent. You reach for it when you are past the prototype stage and need to produce or refine video for distribution—product demos, course clips, shorts, or webinar trims—without living inside a desktop NLE for every small change. It spans multiple journey moments: validating a narrative with a rough cut, polishing assets during build-related docs, and compounding reach in grow/content. Expect intermediate familiarity with codecs, aspect ratios, and source files; the skill orchestrates steps rather than replacing creative direction. Pair it with your local ffmpeg or editor toolchain and explicit file paths so the agent can propose commands and edit plans safely.795installs2Security AuditorSecurity-auditor is an agent skill packaged as an OWASP Top 10 2021 reference guide for indie developers shipping web APIs and SaaS backends. It explains what each top-ten category means in plain language, shows common failure modes such as missing auth middleware, insecure direct object references, privilege escalation via parameters, JWT tampering, and CORS misconfiguration, and pairs each area with detection snippets you can compare against your codebase. The intended workflow is not a one-click scanner but guided review: you or your coding agent inspects endpoints and handlers using the patterns, then applies the numbered remediation practices like explicit authorization, ownership checks on object access, and alerting on repeated access-control failures. Use it when you are building new authenticated routes, refactoring authorization, or doing a pre-ship security pass without a dedicated AppSec team. It fits Claude Code, Cursor, and similar agents that can read repository files and reason over diffs. Complexity is intermediate because you need enough backend context to map patterns to your stack. Treat output as engineering guidance to validate with tests and your own threat m1installs