v6.4.2
obra/superpowersv6.4.2Sep 25, 2026by obra
Upgrading?
AI Summary
The `writing-plans` skill has been optimized to produce leaner, faster plans by recording decisions like signatures and test assertions instead of writing out full code. This change prevents frontier models from becoming overzealous and ensures plans remain focused on the design phase.
Key Highlights
- Plans now record decisions (signatures, test assertions, spec values) rather than writing out the full code implementation.
- The new skill prevents frontier models (like Opus 5.5) from getting overzealous and trying to implement the entire project during the planning phase.
- Plans are now approximately 25% faster and use about 33% fewer tokens compared to previous iterations.
- Self-review checks the proportion of the plan length relative to the spec to ensure it remains a design document.
Breaking Changes
- Removed `plan-document-reviewer-prompt.md`.
- Removed `CLAUDE.md` (replaced by `AGENTS.md`).
New Features
- New 'What a Step Contains' structure defining test steps, code steps, and verification steps.
- Self-review mechanism that checks the proportion of the plan against the spec.
- Updated reader description to reflect a capable engineer rather than zero context.
Full Release Notes
`writing-plans` produces leaner plans, faster. Plans now record the decisions an implementer needs (signatures, test assertions, the spec's values) instead of writing out the code. Some frontier models, including Opus 5.5, could get overzealous during plan writing and, with certain prompting, would sometimes try to implement the entire project while designing the plan. The new skill keeps planning focused on the plan. When we reproduced the original report, the scratch builds went away, and plans took a quarter of the time and about a third of the tokens. Thanks to Harper Reed for the report and session bundle. (#2333) ### Writing Plans - **A plan records decisions. It's not a transcript of the code.** "What a Step Contains" replaces the "No Placeholders" section. A test step names the test and its assertions. A code step gives the exact signature, the file, and the spec's values, and includes a body only for an algorithm those don't determine. A verification step gives the command and its passing output. A reference to another task goes through that task's Interfaces block. Placeholders are still called out as the opposite failure. (#2333) - **Self-review checks proportion.** The plan compares its own length to the spec's. A plan several times longer than the spec is a transcript, and when code blocks dominate, bodies get replaced with signatures and test assertions. (#2333) - **The plan's reader is described as capable:** an engineer who writes idiomatic code once they know the exact interface and test. This replaces "zero context, questionable taste." Steps are now sized as "one action with a checkable result" instead of "2-5 minutes." (#2333) - Every plan written by the new skill executed 9/9 against planted-defect probes on Sonnet 5, the same result as full-code plans. (#2333) - Removed `plan-document-reviewer-prompt.md`. Nothing referenced it. (#2333) ### Documentation - Removed `CLAUDE.md`. Claude Code now reads `AGENTS.md` directly, but only when no `CLAUDE.md` exists, so keeping the one-line pointer would have hidden the real guidelines.
Related Repositories
A complete AI agency at your fingertips - From frontend wizards to Reddit community ninjas, from whimsy injectors to reality checkers. Each agent is a specialized expert with personality, processes, and proven deliverables.
Structured multi-perspective deliberation for hard decisions. Run full councils, focused triads, or duo debates across Claude Code, Codex, Gemini CLI, and OpenCode.
Script that installs/configures a Dovecot, Postfix, Spam Assassin, OpenDKIM Debian web server
The repository for the Angular Webpack Cookbook Gitbook
[CVPR 2024] FaceTalk: Audio-Driven Motion Diffusion for Neural Parametric Head Models