tracking-format
Tracking-file frontmatter and coordination format guidance. Use when creating or reviewing local tracking files and related handoff conventions. DO NOT USE FOR: session-memory plan authoring rules outside the documented references here.
- License 许可
- MIT
- 最近更新
- 2026-08-16
- Stars
- ★ 3
- Forks
- 0
01 前言 Front Matter
- name
- tracking-format
- description
- Tracking-file frontmatter and coordination format guidance. Use when creating or reviewing local tracking files and related handoff conventions. DO NOT USE FOR: session-memory plan authoring rules outside the documented references here.
03 SKILL.md 正文
Tracking Format Instructions
Purpose
This document defines the standard YAML frontmatter format for tracking files stored in .copilot-tracking/. These files help agents maintain context and state across work sessions.
Directory Structure
.copilot-tracking/
├── research/
│ └── {date}-{topic}.md
├── archived/
│ └── issue-001-feature-name.md
└── calibration/
└── review-data.json
calibration/— persistent per-finding calibration cache (populated bywrite-calibration-entry.ps1orbackfill-calibration.ps1)review-data.json— JSON calibration data (schema:{ calibration_version: 1, entries: [{ pr_number, created_at, findings[], summary }] })
Note:
.copilot-tracking/stores research notes, archived tracking files, and persistent calibration data. Plans are now stored in session memory at/memories/session/plan-issue-{ID}.md, not as local files. Survival:.copilot-tracking/artifacts arewithin-worktreestate underSMC-13; they survive in the current checkout and machine, but are not durable cross-agent handoffs unless copied to GitHub issues or committed docs.
YAML Frontmatter Format
Scope: The following format applies to
.copilot-tracking/research and tracking files only — not to session memory plan files. For session memory plan YAML fields, see Issue-Planner Section 6 inagents/Issue-Planner.agent.md.
All tracking files MUST include YAML frontmatter at the top of the file:
---
status: in-progress
priority: p1
issue_id: "001"
tags: [feature, api, backend]
created: 2025-12-09
updated: 2025-12-09
---
Field Definitions
Required Fields
| Field | Type | Description |
|---|---|---|
status |
string | Current work status (see Status Values below) |
priority |
string | Work priority level (see Priority Levels below) |
issue_id |
string | Reference to issue number or tracking ID |
created |
date | Creation date in YYYY-MM-DD format |
Optional Fields
| Field | Type | Description |
|---|---|---|
tags |
array | Categorization tags (e.g., feature, bugfix, refactor) |
updated |
date | Last update date in YYYY-MM-DD format |
assigned_agent |
string | Current agent working on this item |
blocked_by |
string | Dependency or blocker reference |
related_issues |
array | Related issue IDs |
Status Values
Use one of these standardized status values:
pending- Work not yet started, awaiting resources or dependenciesin-progress- Actively being worked oncomplete- Work finished and verifiedblocked- Cannot proceed due to external dependencyon-hold- Paused by decision, may resume later
Priority Levels
Use one of these standardized priority levels:
p1- Critical/Urgent - Immediate attention requiredp2- High - Important but not blockingp3- Normal - Standard priority workp4- Low - Nice-to-have, can be deferred
GitHub Label Mapping
When deriving priority from a GitHub issue label, use this mapping:
| GitHub label | Frontmatter value |
|---|---|
priority: high |
p1 |
priority: medium |
p2 |
priority: low |
p3 |
| (unlabeled) | p2 (default) |
File Naming Convention
Format: issue-{ID}-{short-description}.md
Examples:
issue-001-authentication-system.mdissue-042-fix-memory-leak.mdissue-123-refactor-database-layer.md
Content Structure
After the YAML frontmatter, include:
- Summary - Brief description of the work
- Context - Background information and decisions
- Progress - Current state and completed items
- Next Steps - What needs to happen next
- Notes - Any relevant observations or blockers
Example Tracking File
---
status: in-progress
priority: p2
issue_id: "042"
tags: [bugfix, performance]
created: 2025-12-09
updated: 2025-12-09
assigned_agent: "Code-Smith"
---
# Issue #42: Optimize Query Performance
## Summary
Database queries in the reporting module are taking too long. Need to add indexes and optimize query structure.
## Context
- User reported 30+ second load times
- Profiling identified N+1 query problem
- Decision: Add composite index on frequently queried columns
## Progress
- [x] Profiled slow queries
- [x] Identified missing indexes
- [ ] Apply index migrations
- [ ] Validate performance improvement
## Next Steps
1. Create migration for new indexes
2. Test in staging environment
3. Measure performance improvement
4. Update documentation
## Notes
- Be careful with index size - monitor disk usage
- Consider query caching as future enhancement
Archiving Completed Work
For any tracking files (research notes, prompt output) in .copilot-tracking/ that reach complete status:
- Move the file to
.copilot-tracking/archived/ - Update
statustocomplete - Add
completeddate field - Keep file for historical reference
Plans saved to session memory (/memories/session/plan-issue-{ID}.md) do not need archiving — session memory is scoped to the conversation.
Cloud Agent Handoff Protocol
This section is retired as a plan/design/session-memory handoff source of truth and delegated to the canonical contract at skills/session-memory-contract/SKILL.md (SMC-01, SMC-03, SMC-08, SMC-13); that delegation retires the old cloud handoff table here. Tracking-format only governs .copilot-tracking/ file frontmatter and local tracking-file structure.
Code-Conductor/D9 owns durable handoff persistence. Continue implementation stays same-session and session-memory-only, with no redundant GitHub issue comments. Stop, Pause, or resume later paths persist or append durable GitHub issue comments for <!-- plan-issue-{ID} --> and <!-- design-issue-{ID} --> when those artifacts are needed. When the session-memory plan carries slice_comment_id (authored under the split plan/frame-slices shape), a durable re-emit must preserve that sibling structure — do not inline frame-slice blocks back into the plan comment.
.copilot-tracking/ files are local scaffolding for research notes, prompt output, archived tracking, and calibration artifacts in the same worktree. They are not suitable for durable cross-agent or cloud-agent handoffs where a new branch or fresh checkout is created. Use the session-memory contract for plan/design/session-memory handoff rules, and use GitHub issues, GitHub comments, or committed docs when a handoff must be durable.
Customization
This format is a template. Projects may add custom fields as needed:
test_coverage- Percentage or statusreview_status- Code review statedeployment_target- Environment informationcustomer_impact- Business impact notes
Maintain consistency within your project's tracking files.
Gotchas
| Trigger | Gotcha | Fix |
|---|---|---|
| Writing a new tracking file | Session-memory plan rules get mixed into local .copilot-tracking/ files |
Use this format only for local tracking files and keep plan state in session memory |
04 同仓库更多 Skills
webapp-testing
Playwright end-to-end testing guidance for web apps, with practical patterns and setup steps. Use when creating or improving browser-based E2E coverage, test stability, and CI execution. DO NOT USE FOR: React component tests (use ui-testing), canvas game interaction (use browser-canvas-testing), TDD workflow (use test-driven-development), debugging failures (use systematic-debugging), or randomized property tests (use property-based-testing).
verification-before-completion
Evidence-based verification checklist before marking work complete. Use before PRs, releases, marking tickets done, or any "I'm finished" declaration; also when naming a baseline commit for a differential suite claim, when designing a mutation campaign or red-run evidence, when making an absence claim, a universal claim, or a one-time demonstration offered as a standing guarantee, when attributing a measurement or claiming a sweep or verification loop is complete, and when stating what a check or a CI suite will do about a change. DO NOT USE FOR: post-merge cleanup or archival (use post-pr-review) or processing GitHub review comments (use code-review-intake).
validation-methodology
Reusable validation and review methodology for staged validation, failure triage, and prosecution-depth setup. Use when running validation ladders, triaging failures, or executing adversarial review passes. DO NOT USE FOR: CE Gate orchestration, specialist dispatch ownership, or step execution flow (keep those in Code-Conductor).
upstream-onboarding
Shared opening-phase protocol for upstream agents (Experience-Owner, Solution-Designer, Issue-Planner) and Code-Conductor when invoked on an existing GitHub issue. Renders a scaled context brief and runs a standards check on inherited work at each phase boundary. Use when a user-invocable agent receives a request referencing an existing GitHub issue and must load solution-authoring and upstream-onboarding at the opening phase. DO NOT USE FOR: subagent dispatches (which already operate within an assessed session context); post-merge review (use post-pr-review); research or non-tree subagents.
ui-testing
Resilient React component testing strategies focusing on user behavior. Use when writing or reviewing component-level React tests, fixing flaky component tests, or establishing React testing patterns. DO NOT USE FOR: Playwright E2E tests (use webapp-testing), canvas game interaction (use browser-canvas-testing), TDD workflow (use test-driven-development), debugging failures (use systematic-debugging), UI component design (use frontend-design), or randomized property tests (use property-based-testing).
ui-iteration
Reusable screenshot-driven UI polish workflow for iterative visual refinement. Use when improving layout, hierarchy, spacing, feedback, or Tailwind-based presentation through repeated review-and-adjust passes. DO NOT USE FOR: functional bug fixing, accessibility-only audits, or major redesign strategy work (use Code-Smith or frontend-design).