{"slug": "a-tldr-command-for-claude-code-review-and-summarize-session-work-as-clean-great", "title": "A /tldr command for Claude Code: review and summarize session work as clean Markdown. Great for parallel/multi-session workflows.", "summary": "An engineer introduced a /tldr command for Claude Code that reviews and summarizes session work as clean Markdown, designed for parallel and multi-session workflows. The command uses git log, git status, and git diff to ground the summary in recent commits and working tree changes, while relying on conversation context for intent and outstanding items. It outputs structured Markdown with sections for overview, tickets, commits, decisions, and manual tasks, omitting empty sections.", "body_md": "| description | Summarize the work performed during the current session |\n|---|---|\n| allowed-tools | Read, Bash(git log:*), Bash(git status:*), Bash(git diff:*) |\n\n- Recent commits: !\n`git log --oneline -30`\n\n- Working tree status: !\n`git status --short`\n\nReview the work performed during the current session and summarize it for the user in Markdown, using the structure below.\n\nScope \"the current session\" to work done in this conversation. Use the git context above to ground the Commits section; rely on the conversation for intent, tickets, and outstanding items. Omit any section with no relevant content — no empty headings, no \"N/A\" placeholders.\n\nBroad-strokes bullets covering the most important work done this session. Omit technical detail unless it is critical or poses a risk to the system once deployed.\n\nTickets and epics created for this project during the session, shown in hierarchical\norder: each epic on its own line with its child stories nested beneath it. Render as\na tree, using `├─`\n\nfor each child and `└─`\n\nfor the last child under an epic. List\nstandalone tickets (no parent epic) at the top level with no prefix. Wrap the whole\nblock in a fenced code block so the tree alignment is preserved. For example:\n\n```\nEPIC-123 — Build the resume scoring pipeline\n  ├─ STORY-456 — Implement the ATS parser\n  ├─ STORY-457 — Define the scoring rubric\n  └─ STORY-458 — Wire up the JSON output schema\n\nEPIC-124 — Investor summary refresh\n  └─ STORY-461 — Tighten unit-economics assumptions\n\nTASK-999 — One-off cleanup with no parent epic\n```\n\nCommits made during this session, each as `<short-sha> — <subject>`\n\n.\n\nDecisions still needed from the user.\n\nManual tasks the user must perform, in the order they should be done.", "url": "https://wpnews.pro/news/a-tldr-command-for-claude-code-review-and-summarize-session-work-as-clean-great", "canonical_source": "https://gist.github.com/rcracel/d275778a65de9c9f003be7387d9308d6", "published_at": "2026-06-26 20:24:10+00:00", "updated_at": "2026-06-26 22:33:42.592241+00:00", "lang": "en", "topics": ["developer-tools", "large-language-models", "ai-tools"], "entities": ["Claude Code", "Markdown", "git"], "alternates": {"html": "https://wpnews.pro/news/a-tldr-command-for-claude-code-review-and-summarize-session-work-as-clean-great", "markdown": "https://wpnews.pro/news/a-tldr-command-for-claude-code-review-and-summarize-session-work-as-clean-great.md", "text": "https://wpnews.pro/news/a-tldr-command-for-claude-code-review-and-summarize-session-work-as-clean-great.txt", "jsonld": "https://wpnews.pro/news/a-tldr-command-for-claude-code-review-and-summarize-session-work-as-clean-great.jsonld"}}