copilot-tracking — правило для Claude Code
Правило задаёт, как хранить промежуточные материалы RPI, HVE Builder и других рабочих процессов в .copilot-tracking/. Оно помогает агенту вести исследование, планы и ревью в согласованных файлах.
MIT ★ 1 627 обновлён 2026-11-03
.github/instructions/copilot-tracking.instructions.mdЧто делает copilot-tracking
Правило направляет рабочие материалы в .copilot-tracking/ и задаёт для них структуру каталогов и имена файлов. Для исследований, планов, критики планов, изменений и ревью предусмотрены отдельные пути с датой и коротким слагом задачи. Слаг пишется в lower-kebab-case и ограничен 60 символами.
Для планов и записей об изменениях правило требует оформлять команды и код в обратных кавычках, а существующие файлы связывать относительными Markdown-ссылками. В других артефактах пути остаются обычным текстом. Исследовательские выводы и решения должны попадать в основной документ исследования; ссылки на внутренние артефакты нельзя переносить в производственный код и сообщения коммитов.
Чтобы установить правило, сохраните его в .claude/rules/copilot-tracking.md. Оно касается артефактов рабочих процессов, а не файлов приложения; слаг длиннее лимита нужно сократить по заданному правилу.
Какие задачи решает
- сохранить результаты исследования по принятому пути
- разложить планы и ревью по датированным файлам
- связать существующие файлы в плане относительными ссылками
- не перезаписать свидетельства предыдущего запуска HVE Builder
Примеры запросов агенту
- Сохрани результаты исследования задачи
user-authв основном артефакте.copilot-tracking/research/. - Создай план и запиши его в
.copilot-tracking/plans/по правилам для путей и ссылок. - Подготовь запись изменений в
.copilot-tracking/changes/, не добавляя внутренние пути в сообщения коммитов.
Как установить copilot-tracking
Вручную: положить файлы
.github/instructions/copilot-tracking.instructions.mdПоле applyTo задаёт, к каким файлам применять
Вручную: положить файлы
.claude/rules/copilot-tracking.mdПоле paths во frontmatter ограничивает файлы: applyTo из источника → paths
AGENTS.md
Codex читает AGENTS.md: вставьте текст правила туда
Вручную: положить файлы
.cursor/rules/copilot-tracking.mdcВо frontmatter: description, globs, alwaysApply. applyTo из источника → globs, иначе правило само не подключится
Вручную: положить файлы
.devin/rules/copilot-tracking.mdВручную: положить файлы
.clinerules/copilot-tracking.mdПо шагам, с командами для каждой платформы: AGENTS.md и CLAUDE.md: что это, кто читает и какой выбрать
Вопросы о copilot-tracking
Как установить copilot-tracking в Claude Code?
Сохраните правило по пути .claude/rules/copilot-tracking.md.
Что делает copilot-tracking?
Задаёт правила хранения и оформления промежуточных артефактов RPI, HVE Builder и связанных рабочих процессов в .copilot-tracking/.
Какой формат имени задачи использовать?
Используйте lower-kebab-case длиной не более 60 символов. Если слаг длиннее, сократите его до последнего дефиса в пределах лимита либо до самого лимита, когда такого дефиса нет.
Исходный текст
copilot-tracking.instructions.md · MIT
---
description: "Shared .copilot-tracking conventions for RPI, HVE Builder, proposal response, and compatibility workflow evidence"
applyTo: '.copilot-tracking/research/**, .copilot-tracking/plans/**, .copilot-tracking/changes/**, .copilot-tracking/reviews/**, .copilot-tracking/rpi-sessions/**, .copilot-tracking/challenges/**, .copilot-tracking/prompts/**, .copilot-tracking/walkthroughs/**, .copilot-tracking/hve-builder/**, .copilot-tracking/proposal-responses/**'
---
# Copilot Tracking Conventions
Apply these conventions whenever an RPI, HVE Builder, or compatibility workflow writes intermediate, working, or scratch artifacts under `.copilot-tracking/`.
## Core Rules
* Default to `.copilot-tracking/` for every intermediate, working, or scratch file a skill produces. This file-based tracking takes precedence over memory: persist durable working state to the dated tracking artifact rather than relying on session, conversation, or working memory.
* Resolve where `.copilot-tracking/` lives and search its gitignored files as `copilot-tracking-location.instructions.md` defines.
* Persist research, planning, details, changes, and review outputs under `.copilot-tracking/` using the conventions below.
* Use `{{task_slug}}` for task slugs and `{{YYYY-MM-DD}}` for dates. Keep `{{task_slug}}` lower-kebab-case and at most 60 characters, because it names every tracking file; shorten a longer slug at the last hyphen within the cap, or at the cap when there is none.
* Generated tracking Markdown starts with `<!-- markdownlint-disable-file -->` and never uses `#file:` directives or line-number references.
* Plans under `.copilot-tracking/plans/` and changes records under `.copilot-tracking/changes/` are read by people in an editor. In those two artifacts, wrap code, commands, and symbols in backticks, and link an existing file or folder with the workspace-relative path as the link text and a path relative to the artifact file as the destination, for example `[scripts/plugins/Sync-PluginManifest.ps1](../../../scripts/plugins/Sync-PluginManifest.ps1)`. Keep a not-yet-created path in backticks.
* Research, critique, review, and other tracking artifacts use plain-text workspace-relative paths without Markdown links or backticks around paths.
* Markdown is the default tracking format. Versioned payload contracts and session state use the YAML or JSON format defined by their owning schema. Do not add `applyTo` metadata or create a `.instructions.md` tracking artifact.
## Handoff Expectations
* Keep the parent skill response compact and evidence-first. Write full detail to the tracking file that the phase owns.
* When a handoff is required, name the next phase and the expected artifact path instead of inlining the downstream workflow.
## RPI Research Evidence Ownership
* The primary research artifact is the only research artifact. It owns synthesized questions, findings, canonical evidence IDs, current and unresolved decisions, planning readiness, and user research decisions.
* A research helper, when one is used, returns source locations, excerpts, and brief notes in conversation. Those returns are suggestions, not tracking artifacts; the research context verifies them at the source before recording evidence.
* Persist user answers, unanswered questions, resulting decisions, and selected further-research items in the primary research artifact before the next research action.
## Tracking File Conventions
* Primary research notes stay under `.copilot-tracking/research/{{YYYY-MM-DD}}/{{task_slug}}-research.md`. When the caller supplies a trusted alternate evidence root, they stay at `<root>/research/{{YYYY-MM-DD}}/{{task_slug}}-research.md`, with the date resolved as `rpi-research` defines.
* Planning evidence stays under `.copilot-tracking/plans/{{YYYY-MM-DD}}/{{task_slug}}-plan.md`.
* Plan critique evidence stays under `.copilot-tracking/reviews/plans/{{YYYY-MM-DD}}/{{task_slug}}-plan-critique.md`. A follow-up critique of the same plan adds a numbered suffix, such as `{{task_slug}}-plan-critique-2.md`.
* Implementation evidence stays under `.copilot-tracking/changes/{{YYYY-MM-DD}}/{{task_slug}}-changes.md`.
* Review evidence stays under `.copilot-tracking/reviews/logs/{{YYYY-MM-DD}}/{{task_slug}}-review.md`. A new review of the same work adds a numbered suffix, such as `{{task_slug}}-review-2.md`.
* A session that takes on a later request names that work's documents for the work instead of numbering them: a short lower-kebab-case name that describes the work and differs from the session's other document names, such as `task-board-polish`, takes the place of `{{task_slug}}` in its research and plan paths. The plan's critique, changes record, and review take the plan's name. The RPI Agent's state lists every document and names the ones the current work uses, and with the documents' references to each other, it ties a session's documents together.
* RPI Agent session state stays under `.copilot-tracking/rpi-sessions/{{YYYY-MM-DD}}/{{session_slug}}/{{task_slug}}-state.json`. One folder holds every task of one session and is dated by the session's first day, and `{{session_slug}}` is at most 40 characters. The RPI Agent's State contract defines how session folders and state file names are derived, shortened, and kept unique, and defines the state fields; research, plan, critique, changes, and review records keep their own dated folders.
* Challenge session records stay under `.copilot-tracking/challenges/{{YYYY-MM-DD}}/{{task_slug}}-challenge.md`.
* Walkthrough decisions-and-changes ledgers stay under `.copilot-tracking/walkthroughs/{{YYYY-MM-DD}}/{{task_slug}}-decisions.md`.
* HVE Builder stage evidence stays under `.copilot-tracking/hve-builder/{{YYYY-MM-DD}}/{{artifact_slug}}-{{stage}}-{{attempt}}.md`. Scan existing files and increment `{{attempt}}` rather than overwriting another run.
* Proposal-response evidence stays under `.copilot-tracking/proposal-responses/{{response_slug}}/response-evidence.yml`. Analyze, contribute, and draft operations update this canonical artifact in place while preserving stable record IDs; requested renderings use stable sibling filenames.
* Keep `.copilot-tracking/` paths and other internal planning, research, or implementation artifact references out of production code, code comments, documentation strings, and commit messages. Internal artifacts guide implementation logic; comments stay self-contained and may cite public materials such as RFCs, specifications, or official documentation.
* For the research phase, keep writes inside `.copilot-tracking/research/`, or inside the caller's trusted alternate evidence root when one is supplied, except for workflow tracking files that the current execution explicitly requires.
* When material gaps remain, re-enter the current phase and update the dated artifact rather than skipping ahead. A user-directed skip of an optional step, such as the plan critique or review, is not a material gap.
## RPI Identity and Marker Conventions
* Use one stable task ID across the plan, critique, changes, and review records.
* Use `Pxx` for phases and `Pxx-Txx` for tasks. Place `<!-- rpi:phase id=P01 -->` or `<!-- rpi:task id=P01-T01 -->` immediately before the matching heading.
* Use descriptive headings and related plan or task markers for implementation evidence. Do not create a second per-entry identity scheme in the changes record.
* Use `PC-xxx` only in the plan critique and `RV-xxx` only in the review record.
* Use stable IDs, markers, and headings to navigate. Do not maintain line numbers, line ranges, or detail-line verification.
© microsoft · MIT · microsoft/hve-core