3be9b17, the version that was reviewed and scanned here. Newer commits reach this catalog only after a new review. License: MIT, by Off Grid AI.
Engineering
- Follow the project's contract. Find its engineering rules first: the agent instructions at the repo root
(
AGENTS.md,CLAUDE.mdor similar), aCONTRIBUTING.md, or an engineering contract the instructions point to. Changes are additive. No unrelated refactors. Ask the user before adding a new abstraction, schema or migration. - Code first, check once. Finish the change and review the diff. Then run typecheck, lint and the tests that touch the changed files once, in the background. Take the real commands from the package scripts or CI config, not from memory. Don't chase optional dashboard tools (code-quality bots and the like) unless the project gates on them.
- Small commits. One change per commit. Stage named files only, never
git add -Aorgit add .. Use the project's branch naming (for example git-flowfeature/...), and the same branch name in every repo the change spans. - Push with hooks. Never
--no-verify. If the repo uses submodules and you committed inside one, commit the new submodule pointer in the parent repo too and push both, submodule first. - No worktrees unless asked. If the user asks for one, run
lean-worktree.sh <dir>from thebe-my-ctoskill folder before installing dependencies, so the worktree sharesnode_moduleswith the main checkout. - Never wait. Long jobs (builds, full test runs, installs) run in the background. Carry on and resume when they finish.
Engineering
- Follow the project's contract. Find its engineering rules first: the agent instructions at the repo root
(
AGENTS.md,CLAUDE.mdor similar), aCONTRIBUTING.md, or an engineering contract the instructions point to. Changes are additive. No unrelated refactors. Ask the user before adding a new abstraction, schema or migration. - Code first, check once. Finish the change and review the diff. Then run typecheck, lint and the tests that touch the changed files once, in the background. Take the real commands from the package scripts or CI config, not from memory. Don't chase optional dashboard tools (code-quality bots and the like) unless the project gates on them.
- Small commits. One change per commit. Stage named files only, never
git add -Aorgit add .. Use the project's branch naming (for example git-flowfeature/...), and the same branch name in every repo the change spans. - Push with hooks. Never
--no-verify. If the repo uses submodules and you committed inside one, commit the new submodule pointer in the parent repo too and push both, submodule first. - No worktrees unless asked. If the user asks for one, run
lean-worktree.sh <dir>from thebe-my-ctoskill folder before installing dependencies, so the worktree sharesnode_moduleswith the main checkout. - Never wait. Long jobs (builds, full test runs, installs) run in the background. Carry on and resume when they finish.
Scanned by
NVIDIA SkillSpector
Automated static analysis (no LLM) · v2.12.0 · 2026-11-20
Partial analysis: some checks could not run on this component.
0 observations
SkillSpector looks for patterns such as shell commands, credential access, network calls or unpinned dependencies. An observation is something worth reading in the files before you install, not a confirmed vulnerability. Components that run code locally, like hooks, MCP servers and security tooling, naturally show more of them. Open one to see the lines it points at.
No observations.
Learn more about NVIDIA SkillSpector or how components are reviewed.
Agent compatibility
Checked by NVIDIA SkillEvaluator Harness Check · v1.0.1 · 2026-11-20
How each agent reads this skill once it is installed, from the agents' documented loading rules. No agent was run. A note means the agent loads the skill but changes part of its text, or drops it.
Claude Code
loadsCodex
loadsCopilot CLI
loadsOpenCode
loadsHermes
loadsOpenClaw
loads
Every agent loads this skill as written. No notes.
Learn more about NVIDIA SkillEvaluator.