skills.rest
Kaidanov
KaidanovCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

tldr

Generate end-of-turn status reports with token usage, time, and work summaries.

2|1|Updated Jun 23, 2026
Search Tags:#claude code#status report#daily log#session summary#powershell#stop hook#token tracking
One-click install
npx skills add https://github.com/Kaidanov/grekai-skills-4all --skill tldr-kaidanov
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: tldr
Source: https://github.com/Kaidanov/grekai-skills-4all/tree/main/skills/tldr
Command: npx skills add https://github.com/Kaidanov/grekai-skills-4all --skill tldr-kaidanov

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve? Long agent sessions make it hard to remember what was asked, what shipped, and how much was spent without re-reading the entire transcript. This Skill produces a scannable end-of-turn status memo so you can decide the next step immediately and track token and time spend. ## Core Features & Use Cases - Structured status report: Emits original ask, work statistics (tokens, time, rounds, tools), reasoning summary, Q&A, done, to-do, issues, options, and a fresh-session recommendation. - Honest measurement contract: Per-subagent tokens and durations are reported exactly from usage blocks, while totals and wall-clock are labeled estimates and unmeasurable counts are marked n/a. - Daily trail log: Bundled PowerShell scripts append reports and per-turn heartbeats to a shared daily markdown log with cross-process locking and real wall-clock timestamps. - Use Case: After a multi-round refactor, type /tldr to get a memo of what shipped, what remains, and whether to start a fresh session, then archive it to the daily trail. ## Quick Start Ask the assistant for a TLDR status report of the work just completed in this session.

Frequently Asked Questions about tldr

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I get a status summary of what my AI agent did?▼

Type /tldr or ask for a status, wrap-up, or TLDR at the end of a work batch. The assistant emits a structured memo covering the original ask, work statistics, what shipped, what remains, issues, and next options.

How do I track token usage per subagent in Claude Code?▼

Per-subagent tokens and durations are read exactly from each subagent's returned usage block and shown in a components table. Total tokens and wall-clock time are not exposed to the model, so they are reported as labeled estimates.

Does the tldr skill work on macOS or Linux?▼

The in-chat report works anywhere with no external tools. The daily trail scripts require PowerShell, which ships on Windows and is available as pwsh on macOS and Linux.

How do I log every turn automatically without typing /tldr?▼

Wire the bundled Log-TurnHeartbeat.ps1 script as a Stop hook in your Claude Code settings.json. Every turn then appends a one-line heartbeat with your prompt gist and wall-clock gap to the daily trail file.

Why are some token numbers marked n/a in the report?▼

Per-tool, per-skill, and per-hook token counts are not itemizable by the harness, so they are marked n/a rather than invented. The skill's hard rule is to never fabricate numbers it cannot measure.

Related Skills

Explore More Software Engineering Skills→
cowwoccowwoc
101

cat:token-report

Generate a deterministic token usage report for AI subagents.

Community
Basic
evolution-foundationevolution-foundation
520

prod-good-morning

Generate a pt-BR morning briefing from calendar, Gmail, and Todoist data.

Official
Advanced
haingt-devhaingt-dev

token-optimize

Audit Claude Code token consumption across context sources and MCP servers.

Community
Advanced
chiptoe-svgchiptoe-svg

add-dashboard

Add a local monitoring dashboard for NanoClaw deployments.

Community
Intermediate