SkillFlux

    ↑ ↓ select↵ openSearch all of SkillFlux
    Browse directory
    Aspectawesome-claude-skills (BehiSecc)Superpowers (obra)
    Main purposeSecurity-oriented skill listPlanning, debugging and testing workflows
    Source typeOpen-source repos · Open-source skill repositoryOpen-source repos · Open-source skill repository
    Use casesSecurity & review, Find the right skillCoding & development, Planning & long tasks, Security & review
    LanguageEnglishEnglish
    PricingFreeFree
    Best forDevelopers concerned with skill safety or seeking candidates for authorized review tasks.Teams bringing planning, debugging and testing into a shared development process. Developers willing to read the framework’s rules and adapt them to a project.
    LimitationsA security label is not an audit conclusion, and the list does not establish that linked scripts are risk-free.Framework rules may overlap with project conventions and need to be reconciled before adoption. Testing and review instructions are not themselves test results or security conclusions.
    SkillFlux sample50% clean · 1/18 fail at least one check · 6/18 do not say when to use the skill · 2/18 have no license we could find100% clean · 0/15 fail at least one check · 0/15 do not say when to use the skill · 0/15 have no license we could find

    awesome-claude-skills (BehiSecc): SkillFlux note

    Define the review target and authorized scope before inspecting what a linked project executes. Treat candidates as tools to validate: check whether their findings can be reproduced and whether the output supplies enough evidence to locate a problem.

    Visit awesome-claude-skills (BehiSecc) ↗

    Superpowers (obra): SkillFlux note

    Start with a change covered by existing tests and inspect the output of each step from planning to verification. The process should serve the project’s acceptance criteria. Explain adjustments where a step does not fit; completing a process is not proof that the feature works.

    Visit Superpowers (obra) ↗

    "Clean" is the share of sampled entries with no failing check, a "when to use" description and a license. Sampling reads files only and does not show whether a skill works. Method