claude

aisa

Skills, commands, and hooks for creating and maintaining AI-ready project configurations. Provides scaffolding for CLAUDE.md, .claude/ directory structure, skills, commands, and hooks.

2 on GitHubby rnagrodzkiv0.9.0MITupdated 7 months ago
/

Plugin aisa from marketplace ai-setup-automation in rnagrodzki/ai-setup-automation. Commands read from the repo's manifest — 1 month ago.

This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.

What it installs

Skills (11)

  • aisa-audit

    Use this skill when you need a deep read-only review of .claude/ content accuracy. Triggers: audit AI configuration, review skills accuracy, check principle compliance, find outdated code examples, identify coverage gaps, validate skill specificity. Reports HEALTHY / NEEDS_ATTENTION / CRITICAL.

  • aisa-cache

    Manage .claude/cache/ snapshots for incremental scanning — supports rebuild, status, and invalidate sub-commands. Reduces token consumption by 60-80% by skipping unchanged content. Run after any sync cycle or manually to refresh.

  • aisa-harvest

    Promote ACTIVE learnings from .claude/learnings/log.md into skill gotchas, new skills, and documentation — marks entries as PROMOTED or STALE. Run when learning log has 10+ ACTIVE entries.

  • aisa-init

    Use this skill when initializing or rebuilding AI-assisted development configuration for a project. Scans the project tech stack and generates a complete .claude/ setup (CLAUDE.md, skills, agents, learnings, cache) from scratch.

  • aisa-inspect

    Quick read-only drift check — runs snapshot + drift audit phases only, auto-fixes only CRITICAL issues with approval. Use weekly or before sprints.

  • aisa-lint

    Lint skills/agents for structural completeness and principle compliance (P1-P3, A1-A6) — does not check codebase accuracy. Use after introducing new skills/agents, or as a pre-flight check before committing.

  • aisa-postmortem

    Use this skill when running a post-mortem after an incident, debugging session, production bug, or any situation involving a hotfix, outage, or revert. Turns incidents into prevention — root-cause-to-skill-gap analysis, learning entries, and targeted skill updates.

  • aisa-principles

    Shared architectural principles, tool registry, and behavioral rules for all aisa-* skills. Not invoked directly — loaded as a dependency by other skills.

  • aisa-spec-check

    Check openspec CLI availability, project initialization, and version currency. Suggests installation, initialization, or update with user confirmation.

  • aisa-sync

    Full 7-phase sync cycle — snapshot, drift audit, learnings harvest, expansion analysis, change plan, critique, and execute. Keeps .claude/ in sync with the current codebase. Run every 2-4 weeks or after major features.

  • aisa-update

    Scoped update of skills/agents after a specific change — uses git diff to limit scope, flags out-of-scope drift without fixing it. Faster than full sync. Use after shipping a feature or completing a refactor.

Hooks (1)

  • SessionStart

    command -v openspec >/dev/null 2>&1 && echo "openspec: v$(openspec --version 2>/dev/null) installed$([ -d .openspec ] && echo ', project initialized' || echo ', project NOT initialized — run /aisa-spe

View source on GitHub

Maintain aisa?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex pluginslisted