Skip to content

get-fable v1.4.0 — Canonical Skill Pack, Multi-Agent Matrix & Universal Playbooks

Choose a tag to compare

View all tags
@github-actions github-actions released this 28 Aug 08:00
· 13 commits to master since this release
v1.4.0
11914be

get-fable v1.4.0 — Canonical Skill Pack, Multi-Agent Matrix & Universal Playbooks

get-fable v1.4.0 delivers a major upgrade to the AI agent coding lifecycle: deepening all 25 canonical skills into production-ready operational playbooks, adding native marketplace integration and authentic SVG assets across 10 agent harnesses, and hardening workspace/evidence isolation boundaries.

🌟 Key Highlights

1. Complete Operational Playbooks for All 25 Canonical Skills

Every canonical skill has been upgraded with structured trigger conditions, deterministic step-by-step procedures, tool usage policies, evidence requirements, and recovery flows:

  • Core Lifecycle: get-fable, fable-discover, fable-plan, fable-execute, fable-verify, fable-recover
  • Intelligence & Build: fable-research, fable-tdd, fable-delegate
  • Quality & Safety Gates: fable-review, fable-security, fable-eval
  • Delivery & Orchestration: fable-release, fable-handoff, fable-cowork, fable-loop
  • Execution & Simulation: fable-run, fable-simulator, fable-spark, fable-simplify
  • Context & Memory: fable-artifact, fable-dataviz, fable-memory, fable-config, fable-mode

2. Multi-Agent Ecosystem & Native Marketplace Support

  • Native plugins, rules, and configuration matrices for 10 top agent platforms:
    • Claude Code (.claude-plugin/)
    • OpenAI Codex (.codex-plugin/)
    • ChatGPT (.chatgpt-plugin/)
    • Gemini CLI / Antigravity (.gemini-plugin/)
    • Cursor (.cursor-plugin/)
    • OpenCode (.opencode-plugin/)
    • DeepSeek (.deepseek-plugin/)
    • Kimi (.kimi-plugin/)
    • Kiro (.kiro-plugin/)
    • Pi Code (.pi-plugin/)
  • Unified single-command installer matrix for all platforms (get-fable install <agent>).
  • Authentic high-resolution vector SVG logos and UI assets integrated across all supported agents and hosts.

3. Enterprise-Grade State & Verification Boundary Hardening

  • Bound evidence records strictly to workspace ID and routing scope to eliminate cross-workspace bleed.
  • Implemented failure precedence preservation in evidence gates.
  • Added native Git lifecycle hook execution in linked worktrees.
  • Enforced strict supply-chain pinning across all GitHub Actions workflows with full commit SHAs and Bun frozen lockfiles.

📦 Quick Start & Installation

# Direct run with Bun
bunx get-fable init

# Install into your favorite coding agent
bunx get-fable install claude
bunx get-fable install antigravity
bunx get-fable install codex
bunx get-fable install cursor
bunx get-fable install opencode

📋 Full Changelog

  • feat(skills): Deepen all 25 canonical Skills into operational playbooks (#19)
  • feat(skills): Upgrade get-fable adapter skill to full lifecycle routing guide and extract session learnings
  • feat(assets): Integrate official AI coding agents & IDEs SVG logos via @lobehub/icons
  • feat(assets): Add get-fable infographic GIF & high-res branding assets
  • fix(state): Bind evidence to workspace and routing scope (#22)
  • fix(hooks): Support Git hooks in linked worktrees (#24)
  • fix(state): Invalidate evidence after failed writes and preserve failure precedence
  • ci: Enforce generated llms.txt context and package catalog freshness (#23)
  • ci: Pin supply-chain actions with immutable commit SHAs and OIDC trusted publishing

Full Diff: v1.3.0...v1.4.0

Assets 2
Loading