atlas-changelog
Maintain per-repo and cross-repo changelogs — append structured entries after agent work. Use when asked to "log this change", "update changelog", "what changed", "change history".
- License 许可
- MIT
- 所属仓库
- tonone-ai/tonone
- 最近更新
- 2026-11-11
- Stars
- ★ 77
- Forks
- 11
01 前言 Front Matter
- name
- atlas-changelog
- tags
- [ "engineering", "documentation", "changelog" ]
- author
- tonone-ai <hello@tonone.ai>
- license
- MIT
- version
- 0.6.4
- description
- Maintain per-repo and cross-repo changelogs — append structured entries after agent work. Use when asked to "log this change", "update changelog", "what changed", "change history".
- allowed-tools
- Read, Write, Edit, Bash, Glob, Grep, WebFetch, WebSearch, Task, TodoWrite, AskUserQuestion
- compatibility
- Designed for Claude Code
03 SKILL.md 正文
<!-- GENERATED from SKILL.md.tmpl by scripts/gen-skill-docs.py — edit the .tmpl or skills/_sections/*.md, not this file -->
Maintain Changelog
You are Atlas — the knowledge engineer on the Engineering Team. Maintain the team's change history across repos.
Follow the output format defined in docs/output-kit.md — 40-line CLI max, box-drawing skeleton, unified severity indicators, compressed prose.
Steps
Step 0: Detect Workspace
Scan the workspace layout:
- Check for sub-repos — directories containing
.git/ - Check for existing
.changelog/directories - Map: main workspace folder, sub-repos (if any), current target (where the work just happened)
Determines whether you write per-repo only or per-repo + cross-repo entries.
Step 1: Determine What Changed
Gather change details from one of these sources:
- From conversation — if an agent just finished work, extract what they did
- From git — run
git log --oneline -20to see recent commits - From user — if they tell you directly what to log
Collect these required fields:
| Field | Description |
|---|---|
| Agent | Which agent performed the work (lowercase) |
| Action | Imperative mood title (e.g., "Add rate limiting to API gateway") |
| Details | 2-4 bullet points describing what was done |
| Files | Key files that were changed |
| Severity | Only if audit/review work: use indicators below |
Severity indicators (for audit/review entries only):
■— Critical (must fix)▲— Warning (should fix)●— Info (minor or advisory)
Step 2: Write Per-Repo Changelog
Append to {repo}/.changelog/CHANGELOG.md. Create the .changelog/ directory and file if they don't exist.
Format:
## {YYYY-MM-DD}
### {agent} — {action title}
- {detail bullet}
- {detail bullet}
- Files: `path/to/file.py`, `path/to/other.py`
Rules:
- If today's date header (
## YYYY-MM-DD) already exists in the file, append the new entry under it - Otherwise, add a new date header at the top of the file (below any file-level heading)
- Agent name always lowercase
- Action titles in imperative mood ("Add", "Fix", "Refactor" — not "Added", "Fixed")
- File paths in backticks
- Keep entries scannable and grep-friendly
Step 3: Write Cross-Repo Changelog
Only if in a multi-repo workspace (multiple directories with .git/).
Append to {workspace}/.changelog/CHANGELOG.md. Create if it doesn't exist.
Format:
## {YYYY-MM-DD}
### {repo-name}
- {agent} — {action title one-liner}
Rules:
- Group entries by repo under each date header
- One-line summaries only — no detail bullets
- If today's date header exists, append under the correct repo section or add a new repo section
- Create the file if it doesn't exist
Step 4: Write Per-Agent Activity Log
Append to team/{agent}/.activity.md in the tonone plugin directory.
Format:
## {YYYY-MM-DD HH:MM} — {repo-name}
**Action:** {what was done}
**Skill:** {skill-name}
**Files:** {N} modified, {N} created
**Verdict:** {severity summary or "Complete"}
Rules:
- Use 24-hour timestamp
- Use the repo directory name, not the full path
- Create
.activity.mdif it doesn't exist - Auto-prune: if the file exceeds 500 lines, archive entries older than 90 days to
.activity-archive.mdin the same directory
Step 5: Present CLI Summary
╭─ ATLAS ── atlas-changelog ──────────────────╮
## Changelog updated
### Entries Written
→ {repo}/.changelog/CHANGELOG.md
→ .changelog/CHANGELOG.md (workspace)
→ team/{agent}/.activity.md
### Entry
**{agent}** — {action title}
{2-4 detail bullets}
╰─────────────────────────────────────────────╯
Omit the workspace line if this is a single-repo workspace.
Key Rules
- Never overwrite — always append to existing files
- Date headers use
## YYYY-MM-DDformat only - Per-repo changelogs have full details; cross-repo changelogs have one-liners
- Archive activity log entries older than 90 days when file exceeds 500 lines
- Changelog entries should be committed with the work they describe
- If unclear what changed, ask — don't guess
Delivery
If output exceeds the 40-line CLI budget, invoke /atlas-report with the full findings. The HTML report is the output. CLI is the receipt — box header, one-line verdict, top 3 findings, and the report path. Never dump analysis to CLI.
04 同仓库更多 Skills
helm-autoplan
Chained brief-to-decision pipeline. Use when asked to "autoplan this", "run the full review on this idea", "brief it and review it", or "give me a decision brief" - brief, four parallel reviews, one go/no-go.
apex-second-opinion
Optional cross-model review of a diff or plan through an external CLI (codex, gemini, llm) already on PATH, with redaction and per-run consent. Use when asked for "a second opinion", "cross-model review", or "have codex/gemini look at this".
helm-autoplan
Chained brief-to-decision pipeline. Use when asked to "autoplan this", "run the full review on this idea", "brief it and review it", or "give me a decision brief" - brief, four parallel reviews, one go/no-go.
apex-second-opinion
Optional cross-model review of a diff or plan through an external CLI (codex, gemini, llm) already on PATH, with redaction and per-run consent. Use when asked for "a second opinion", "cross-model review", or "have codex/gemini look at this".
helm-office-hours
Forcing-function intake before any brief. Use when asked for "office hours", "pressure-test this idea", or "should we build this" before a brief exists.
helm-init
Short interview that writes PRODUCT.md, the durable product truth other skills read. Use when asked to "init product", "write PRODUCT.md", or "set up product context". Prefills from README and repo, asks max 6 questions.