openbeast

Author a new task for the eval suite — JSON spec + reference implementation + verified validation. Activate when the user asks to add an eval task, write a benchmark task, or extend the suite. Encodes the 6 hard-won pitfalls from prior post-mortems.

Add multi-language variants (Python / Go / C / C++ / Rust / Zig — 6 supported languages) to an existing eval task. Activate when the user wants to make an eval task language-agnostic or extend the variant rollout. Companion to eval-task-author.

Design a public API contract — function signatures, types, error model, example usage — BEFORE implementing. Activate when adding a new public function / endpoint / library, when changing an existing public surface, or when the user says "design an API for". Frontier models do this naturally; local models often code first and accidentally lock in bad shapes.

Produce a design doc BEFORE writing code for any non-trivial change — new subsystems, public API changes, anything >500 lines or crossing module boundaries. Activate when the user asks for a design, when you find yourself about to invent a non-trivial structure on the fly, or when "let me just start coding" would be premature.

Orient yourself in an unfamiliar codebase BEFORE editing. Activate at the start of any non-trivial task in a repo you don't already know cold — wrong file, wrong style, wrong layer is the most common local-model failure mode and this skill prevents it.

Clean commits, atomic units, meaningful messages, no random staging. Activate any time the user is about to commit, push, or open a PR — local models often stage too aggressively or write messages that don't explain why. Frontier models do this naturally.

Read, structure, and synthesize a large input (10K+ line PR, 30+ files, paper, transcript, log dump) without drowning. Activate when the input doesn't fit in working memory — when "just read it linearly" would consume your context budget without producing structured understanding. This is Kimi 2.6's natural strength encoded as a method.

Measure-driven performance work. Activate when the user asks to make something faster, when you spot what looks like a perf issue, or when adding code in a hot path. Encodes the discipline frontier models follow: profile first, optimize the actual hotspot, re-measure, never speculate.

Extract a precise spec from an open-ended user request BEFORE implementing. Activate at the start of any task where the desired behavior is not fully nailed down — most user requests, in other words. Frontier models pause here naturally; local models often skip this and produce the wrong thing.

Test-driven development discipline — red, green, refactor. Activate when the user asks to TDD a feature, when adding non-trivial functionality to a codebase that has tests, or when the user explicitly wants test-first.

Multi-pass code review for correctness, security, performance, idioms, and tests. Activate when the user asks for review, audit, "look this over", or post-PR analysis.

Systematic root-cause analysis. Activate when something is broken and the obvious fix didn't work — when you've tried two things and need to actually think instead of guessing harder.

Slow-mode reasoning for intractable problems. Activate when the obvious paths have been exhausted, when pattern-matching has failed, or when Max says "consult the council". This is the war council, not autocomplete.

Focused security review — input validation, authentication, cryptography, secrets handling, and timing-safe comparisons. Activate when the user asks for security audit, security review, or to "check for vulnerabilities".