rumoii/rumo-engineering-skills / skills/rumo-coding-guidelines
rumo-coding-guidelines
VerifiedUse when writing, fixing, or refactoring code in a software repository, as the baseline discipline alongside more specific skills. Ground every edit in the real code path, do only the requested work and what it necessarily implies, avoid speculative abstraction and drive-by edits, apply the risk gate to changes touching money, auth, data, interfaces, or other high-risk areas, and close with verification. Do not use alone for review-only, diagnosis-only, or question-answering requests.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 10.1 KB |
| agents/openai.yaml | config | 304 B |
| references/agents-snippet.md | markdown | 2.9 KB |
| references/high-assurance.md | markdown | 9.0 KB |
| references/measurement.md | markdown | 3.1 KB |
| references/principles.md | markdown | 3.9 KB |
| references/task-flows.md | markdown | 6.4 KB |
| scripts/baseline_block.py | script | 5.2 KB |
references/agents-snippet.md
View on GitHub# Always-On Baseline
Clients load a skill only when its description matches the request, so small edits can skip `rumo-coding-guidelines` and one skill may not bring in the next. Keep the block below in a project's or user's `AGENTS.md` or `CLAUDE.md` so every session has the core rules and the routing: run the installer with `-Baseline` / `--baseline`, run `scripts/baseline_block.py --target <file> --install` from this skill, or paste it by hand. It routes to the skills; it does not replace them.
```markdown
## Engineering baseline (rumo skills)
IMPORTANT: Before non-trivial engineering work, such as a feature, bug fix, refactor, debugging, or any multi-step code change, find its row below and load that skill first. General questions and one-line edits need no skill. A subagent dispatched for a specific task follows its brief instead of this routing.
| When | Load |
| --- | --- |
| Any code change beyond a one-line edit | rumo-coding-guidelines |
| Unclear goal, scope, or risky decision | rumo-clarify |
| Agreed multi-step work | rumo-plan |
| Defect or unexplained behavior | rumo-debug, then a failing test before the fix |
| Before saying done, committing, or pushing | rumo-verify |
| A review, or another model's opinion | rumo-review |
| Commit, push, or pull request, only when asked | rumo-git-commit |
| Database writes; deleting files; remote hosts | rumo-database-change-safety; rumo-filesystem-safety; rumo-remote-ops |
| Interfaces other programs read; background work | rumo-interface; rumo-lifecycle-safety |
| Handing work to another agent or model | rumo-agent-collaboration |
| Wrapping up, resuming, or handing off work | rumo-handoff |
| How or why this repository's code works | rumo-explain |
| Technical comparisons or feasibility research | rumo-research |
- Read the real entry point and call chain before editing; repository instructions override these defaults.
- Do only what was asked and what it necessarily implies: every changed line traces to the request, and you report unrelated problems instead of fixing them. No speculative abstractions, drive-by refactors, or reformatting of neighboring code.
- Risk gate before editing: money, auth, secrets, migrations, and irreversible effects are high at once; durable data, concurrency, interfaces other programs read, untrusted input, personal data, deployment, and dependencies are rated first. When high, write a test that fails before the change, prove the one fact the change is safe because of, and rerun the checks after the last edit.
- Never claim done without fresh evidence; use rumo-verify before handing back.
- In answers, explanations, and reports, say where each factual claim comes from: a run you made, code you read (`file:line`), or inference. Never state an inference as fact.
- After repeated user corrections, offer once to review the session for lessons; start only if the user accepts.
- Full rules: the rumo-coding-guidelines skill.
```
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |