ANTIGRAVITY AWESOME SKILLS V10.5.0: GOVERNANCE, SECURITY LABELS, AND OPENCODE STABILITY OVER NEW FEATURES
Antigravity Awesome Skills v10.5.0 tightens governance and security metadata, fixes content issues, and adds OpenCode stability guidance, especially for Windows...
Antigravity Awesome Skills v10.5.0 tightens governance and security metadata, fixes content issues, and adds OpenCode stability guidance, especially for Windows crash/compaction loops.
The v10.5.0 release is a maintenance-heavy update to this installable skills library for Claude Code, Cursor, Codex CLI, and Gemini CLI. The maintainers enforced source-only PRs, closed audit items on main, and expanded governance coverage across COBIT, TOGAF, NIST, and SRE-focused skills.
It adds a README stats sync script (tools/scripts/sync-readme-stats.js via npm run sync-readme), restores a missing Prompt Engineer step, fixes a broken YouTube Summarizer fence, removes stray frontmatter, and backfills date_added. Security labels are normalized, with explicit authorized-use warnings and allowlists for Active Directory, environment setup, and GitOps commands.
Finally, the release ships OpenCode recovery docs with steps for Windows crash and compaction-loop scenarios, aimed at reducing local flakiness. See details in the notes.
Cleaner security metadata and PR hygiene reduce supply‑chain risk and make internal approvals easier.
Stability guidance and bugfixes cut friction in AI‑assisted workflows, especially on Windows.
-
terminal
On a Windows dev machine, follow the OpenCode recovery steps and confirm crash/compaction‑loop issues no longer reproduce.
-
terminal
Run npm run sync-readme and validate skill counts/anchors, then sanity‑check allowlists against your GitOps and AD workflows.
Legacy codebase integration strategies...
- 01.
Pin to v10.5.0, rerun local setup across Claude Code and Cursor, and verify existing skills render correctly after metadata fixes.
- 02.
Map the new risk labels and allowlists to your internal policies to avoid overblocking or gaps during security reviews.
Fresh architecture paradigms...
- 01.
Use the curated bundles and workflows from the release to bootstrap an AI‑assistant toolchain with governance baked in.
- 02.
Adopt source‑only PR enforcement in your repos as a starting guardrail for contributions to skills or templates.
Get daily ANTIGRAVITY + SDLC updates.
- Practical tactics you can ship tomorrow
- Tooling, workflows, and architecture notes
- One short email each weekday