hyhmrright

Scaffold a new brooks-lint analysis skill so it passes `npm run validate` and `npm run evals` on the first try — generates skills/{name}/SKILL.md (with the mandatory "Do NOT trigger for:" clause and a Process section citing guide step ranges) plus skills/{name}/{name}-guide.md (sequentially numbered steps), then appends paired eval scenarios. Triggers when the maintainer asks to "add a new skill", "scaffold a skill", or "create a brooks-lint mode". Do NOT trigger for: editing an existing skill's content, adding a single eval to an existing skill, or authoring skills for some other plugin.

Cut a brooks-lint release: set the version in package.json, propagate it across all four plugin manifests + README badge, write the CHANGELOG entry, validate, then commit, push, tag, and publish the GitHub release. Triggers when the maintainer asks to "release", "cut a release", "ship a new version", or "bump and publish" brooks-lint. Do NOT trigger for: propagating an already-decided version without releasing (use `npm run bump` directly), CHANGELOG edits alone, or questions about the release process that don't ask to perform it.

Full-sweep mode: runs a unified analysis across all quality dimensions — code decay, architecture, tech debt, and test quality — then applies fixes directly to the codebase. Safe changes are auto-applied; risky changes are confirmed before execution. Drawing on twelve classic engineering books. Triggers when: user wants to "fix everything", "sweep the codebase", "auto-fix all issues", "run all checks and fix them", "clean up the whole project", or asks for a single command that both diagnoses and remediates quality problems. Do NOT trigger for: read-only audits or health reports where the user only wants findings without code changes; single-dimension reviews (use the focused skill instead: brooks-review / brooks-audit / brooks-debt / brooks-test); server health checks, HTTP /health endpoints, Kubernetes probes, or application uptime.

Combined codebase health dashboard that scores a project across all four quality dimensions — PR quality, architecture, tech debt, and test quality — in a single pass, drawing on twelve classic engineering books. Triggers when: user wants an overall quality assessment, asks "how healthy is this codebase?", "run all the checks", "give me a big-picture quality report", "I need a health score before the release", "what's the overall state of our code?", or wants to onboard a new team with a quality overview. Do NOT trigger for: server health checks, HTTP health endpoints, Kubernetes liveness/readiness probes, database health, or application uptime. Also do not trigger when the user specifically requests only one dimension — use the corresponding focused skill instead (brooks-review / brooks-audit / brooks-debt / brooks-test).

Architecture audit that maps module dependencies, checks layering integrity, and flags structural decay across a codebase, drawing on twelve classic engineering books. Triggers when: user asks to audit architecture, review folder/module structure, check for circular imports, understand how the codebase is organized, or asks "does this follow clean architecture?", "why does everything depend on everything?", "are our layers correct?", "where should this code live?". Also triggers for onboarding requests: "explain this codebase to a new developer" or "give me a codebase tour" (use onboarding mode). Do NOT trigger for: PR-level code review (use brooks-review) or line-level refactoring questions — this skill analyzes structural/module-level concerns, not individual functions.

Tech debt assessment that identifies, classifies, and prioritizes maintainability problems — helping teams build a refactoring roadmap — drawing on twelve classic engineering books. Triggers when: user asks about tech debt, refactoring priorities, what to clean up first, or asks "why is this so hard to change?", "where's the most painful part?", "what should we fix first?", "how do I justify refactoring to management?", "why is our velocity dropping?". Do NOT trigger for: server health checks, HTTP /health endpoints, Kubernetes probes, database health, or application uptime — "health" in those contexts is infrastructure, not code quality. Also not for single-function refactoring questions.

PR code review that surfaces decay risks, design smells, and maintainability issues with concrete Symptom → Source → Consequence → Remedy findings, drawing on twelve classic engineering books. Triggers when: user asks to review code, check a PR, shares a diff or pastes code asking "does this look right?" / "any issues here?" / "ready to merge?", or asks for feedback on a function, class, or file. Also triggers when user mentions: code smells / refactoring / clean architecture / DDD / domain-driven design / SOLID principles / Hyrum's Law / deep modules / tactical programming / conceptual integrity / Brooks's Law / Mythical Man-Month / second system effect. Do NOT trigger for: questions about how to write code from scratch, language syntax questions, or framework/tool questions where no existing code is shared.

Test quality review drawing on twelve classic engineering books — with primary focus on xUnit Test Patterns, The Art of Unit Testing, How Google Tests Software, and Working Effectively with Legacy Code — that diagnoses structural problems in an existing test suite: brittleness, mock abuse, coverage illusions, slow execution, poor readability. Triggers when: user asks about test quality, shares test files for review, or expresses frustration: "tests keep breaking whenever I change anything", "our tests take forever", "I can't understand what this test is doing", "tests pass but bugs still reach production", "we have too many mocks". Do NOT trigger for: writing new tests from scratch (use the regular test-writing workflow) or testing framework/syntax questions — this skill reviews an existing suite for structural quality problems, not individual test authoring.

Find logic bugs in a single file or function via semi-formal execution tracing (Premises → Trace → Divergence → Trigger → Remedy). Trigger when a user shares code and suspects something is wrong without naming a concrete failure — phrases like "review this", "does this look right", "check this function", "audit this code", "tests pass but prod fails". SCOPE HARD RULE: one file or one function only. For a directory or whole module use logic-health; for a confirmed failure (stack trace, failing test, specific wrong value) use logic-locate; for two versions use logic-diff; for repo-wide autonomous fixing use logic-fix-all. Do NOT trigger for: style/formatting, security scanning, performance, test generation, architecture or design questions.

Compare two code versions for semantic equivalence via semi-formal tracing of both versions side-by-side. Trigger when the user shares a refactor, rewrite, migration, or A/B implementation and wants to confirm behavior is unchanged — "did I break anything", "is this equivalent", "are these equivalent", "semantically equivalent", "are these two implementations semantically equivalent", "check my refactor", "same behavior after the change?", "does my rewrite produce the same output", "switched from X to Y — same results?". SCOPE HARD RULE: requires two code versions (A and B). A single version for bug-finding uses logic-review; one version + a failing test uses logic-locate; explaining what one piece of code does uses logic-explain; codebase audit uses logic-health. Do NOT trigger for: single-version review, performance comparison, design-quality comparison, or "which is better-written" questions.

Sweep a directory, module, or full codebase for logic correctness and produce a scored health dashboard with systemic patterns. Trigger when the user requests a health view — "audit the whole codebase", "health check", "health overview", "logic health overview", "audit src/", "audit auth and payments modules", "where should I focus testing", "onboarding review", "logic overview before we ship", "give me a health overview of this module". SCOPE RULE: prefer multi-file; also trigger for a single module when the user explicitly uses "health check", "health overview", or "logic health" — a concrete failure uses logic-locate; two versions uses logic-diff; explaining a path uses logic-explain; "fix everything" uses logic-fix-all. Do NOT trigger for: style/architecture-only audits, security-only scans, performance-only audits.

Locate the root cause of a CONFIRMED failure via backward-then-forward semi-formal tracing. Trigger when the user provides a stack trace, failing assertion, error message, or specific wrong-value observation — "find the bug", "this test is failing", "track down this crash", "why is this test failing", "KeyError at line 89", "expected 70, got 100", "NoneType has no attribute X", "cart empties when second tab opens". SCOPE HARD RULE: requires a concrete failure (exception, failing test, or specific wrong output). Vague suspicion without evidence uses logic-review; behavior explanation uses logic-explain; refactor comparison uses logic-diff; codebase audit uses logic-health. Do NOT trigger for: vague "what's wrong" without a concrete symptom, style questions, or performance issues.

Explain what a specific piece of code actually does for a given input by producing a step-by-step execution trace (interprocedural, with name resolution and type transitions). Trigger when the user is confused about behavior or asks why code produces X instead of Y — "walk me through this", "trace through X with input Y", "why does this return X", "what does yield-from do here", "explain the execution path". SCOPE HARD RULE: a specific function + a specific input scenario. If the user wants to find bugs without a scenario in mind, use logic-review; two-version comparison uses logic-diff; concrete failures use logic-locate; codebase-wide audit uses logic-health. Do NOT trigger for: finding bugs without a behavioral question, style or design discussion, or concept explanations not tied to specific code.

Autonomous repository-wide audit-and-fix pipeline: health → review → locate/explain → fix → diff-verify → iterate until clean. Starts with a mandatory consent prompt (token-intensive); after consent runs hands-free. Trigger when the user wants ALL logic issues found and fixed — "fix everything", "fix all logic issues", "fix all logic issues in this code", "clean up all logic issues", "audit and fix the whole repo", "fix all bugs automatically", or frustration with recurring bugs wanting a one-shot pass. SCOPE RULE: repo-wide by default; also trigger for a pasted code snippet when the user says "fix all" — produce the full Fix Report (Fix Log, before/after Logic Score). Analysis-only requests use logic-health or logic-review. Single failure uses logic-locate. Two versions uses logic-diff. One path explanation uses logic-explain. Do NOT trigger for: analysis-only ("show me the bugs"), style/lint/format concerns, or fixing a single named finding.

新增 AI Provider，覆盖 config.ts / providers/registry.ts / shared/types.ts 全流程

在 sidecar/agents/masters/ 新增投资大师 Agent，自动完成接口实现和 registry 注册

Summon 13 legendary investors — Warren Buffett, Charlie Munger, Benjamin Graham, Michael Burry, Nassim Taleb, Cathie Wood, and 8 more — to analyze any stock. No API keys. No local setup. Just paste a ticker and go.

Sync plugin snapshot to Git. Use when user wants to sync plugins, push snapshot, or after installing/removing plugins.

Install Codex skills into $CODEX_HOME/skills from a curated list or a GitHub repo path. Use when a user asks to list installable skills, install a curated skill, or install a skill from another repo (including private repos).