using-trove
Skill discipline anchor for the Trove marketplace. Loaded once per session; routes the agent to the right trove-* skill before code, planning, review, or git actions.
- License 许可
- MIT
- 所属仓库
- bravew/trove
- 最近更新
- 2026-11-12
- Stars
- ★ 10
- Forks
- 1
01 前言 Front Matter
- name
- using-trove
- metadata
- { "trove-category": "ai-models-agents", "trove-category-label": "AI models & agents" }
- description
- Skill discipline anchor for the Trove marketplace. Loaded once per session; routes the agent to the right trove-* skill before code, planning, review, or git actions.
- when_to_use
- session start; skill routing; trove bootstrap
- user-invocable
- false
02 资源文件 1
- 其他SKILL.md.tmpl2.4 KB
03 SKILL.md 正文
<!-- AUTO-GENERATED from SKILL.md.tmpl — do not edit directly --> <!-- Regenerate: bun run build:skills -->
using-trove
User instructions outrank Trove skills, and Trove skills outrank default model behavior. When a user instruction conflicts with a skill, ask a concise clarifying question or name the conflict before continuing. Never silently downgrade a user instruction.
Before code changes, planning work, code review, debugging, skill authoring, delegation, or git actions, check whether a trove-* skill applies. If one applies, name the selected skill once in your response, invoke that skill through the available skill-loading mechanism, and follow it. Do not reinvent the skill guidance inline and do not paraphrase this anchor back to the user.
Routing source: read the generated routing fragment routing.md next to the plugin manifest when this plugin is installed. The repository aggregate lives at docs/routing.md; use it as the fallback when the plugin fragment is not present.
When a workflow skill matches, follow its steps in order; if you skip one, say so in a line rather than dropping it silently. When a trove-principle-* skill shaped a decision, name it and the specific choice it changed (a principle cited with no concrete decision behind it was not applied). Run prose you produce — replies, PR text, release notes — through trove-unslop. Proceed on reversible work and present the result for course-correction; still confirm before irreversible or destructive actions per the hard rules below.
Decision gates use the standard format documented in scripts/lib/decision-gate.ts: context, question, options, and default. Link to or invoke the relevant skill instead of duplicating gate wording from another skill.
Hard rules: do not rewrite tracked files without a user prompt, do not run destructive git without explicit confirmation, do not use --no-verify, and do not claim completion until the relevant verification has actually run or the gap is clearly stated.
04 同仓库更多 Skills
trove-write-skill
Skill authoring workflow. Use when creating or updating a Codex or Trove skill, including rubrics, tasks, metadata, and validation.
trove-worktree
Git worktree workflow for spikes, risky experiments, and broad refactors that should not disturb the user's current branch.
trove-why
Reconstruct design rationale and decision history from evidence — source control, issue tracker (Linear), docs (Notion), chat (Slack), error tracking (Sentry) — with calibrated confidence. Use for "why does X work this way", "why did we pick Y", regressions, and postmortems. Use trove-explain for how it works now.
trove-verify
Completion verification workflow. Use before claiming done, fixed, complete, ready, or shipped.
trove-unslop
Cut AI tells from writing — PR descriptions, release notes, specs, commit bodies, and replies. Use before finalizing prose, or when the user asks to clean up, de-slop, or make text sound human.
trove-show-work
Keep a reviewable decision trail for long-running or unattended work: an append-only TSV log with one row per decision (what, why, evidence, result). Use for autonomous or multi-phase runs, or work a human reviews after stepping away.