onboard — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited onboard (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
互動式 wizard,負責建立或更新 company workspace config,並把 company route 登記到 root workspace-config.yaml。
init / /init / re-init 是 deprecated aliases,只作相容入口;新的 skill source、 reference loading 與文件語言都以 onboard 為主。
onboard 只負責 setup orchestration:
_template/workspace-config.yaml 建立或編輯 {company}/workspace-config.yaml。workspace-config.yaml 登記 company routing。_template/rule-examples/ scaffold company-scoped L2 rules。不要猜缺漏值。Auto-detection 只能提供 suggestion,必須經使用者確認。每個 optional section 都可 skip;skip 後欄位保持空值,下游 skills graceful degrade。
依使用者選到的路徑只讀需要的 references:
| Situation | Load |
|---|---|
Any onboard run | onboard-interaction-patterns.md, onboard-core-workflow.md, workspace-config-reader.md, workspace-language-policy.md |
| Existing company / upgrade rerun | onboard-core-workflow.md rerun section |
onboard repair / readiness check | onboard-core-workflow.md, onboard-post-setup-flow.md, workspace-config-reader.md |
| Project repo detection | onboard-interaction-patterns.md, sub-agent-roles.md |
| Dev environment setup | onboard-runtime-setup-flow.md, dependency-consent.md |
| Visual regression setup | onboard-visual-regression-setup.md, visual-regression-config.md |
| Daily learning scanner | learning-setup-flow.md, daily-learning-scan-spec.md |
| Runtime toolchain / Codex bootstrap / handbook | onboard-post-setup-flow.md, repo-handbook.md as needed |
Repo analysis 或 dev environment detection 派 sub-agent 時,必須注入 sub-agent-roles.md 的 Completion Envelope。
Kibana、projects、scrum、infra、runtime、visual regression 設定。
default_company 只能在 root config,不可進 company config。daily learning scanner、root toolchain readiness check、Codex bootstrap、handbook generation。
mise run onboard-doctor,回報 ready / partial / blocked、deferred emptyfields、root toolchain status、Codex bootstrap status 與建議的第一個指令。Root toolchain repair 採 print-only,只顯示 mise run bootstrap / mise run doctor -- --profile <profile>,不可由 onboard repair 隱性代跑。若使用者明確選擇 --execute,才可執行可自動修復且 non-interactive / CI-safe 的 repair command。
print-only,只有明確 --execute opt-in 才能執行 non-interactive / CI-safe command。
發送 user-visible text 前必須通過 workspace-language-policy.md language gate。
產出的 company config 遵守 _template/workspace-config.yaml:
每個 user decision 與 auto-detection result 都 append 到 {company}/.onboard-audit.jsonl;re-run 時 append restart marker。
回報完成前:
post-task-reflection-checkpoint.md checklist。Final response 前執行 shared reflection checkpoint。
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.