Borrowing it
Nothing to install: this file belongs to jackstewart254/groundtruth. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/jackstewart254/groundtruth/main/.claude/skills/tldr/SKILL.mdgit clone --depth 1 https://github.com/jackstewart254/groundtruthWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/jackstewart254/groundtruth/tldr)<a href="https://agentmods.dev/skills/jackstewart254/groundtruth/tldr"><img src="https://agentmods.dev/badge/skills/jackstewart254/groundtruth/tldr/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/jackstewart254/groundtruth/tldr"><img src="https://agentmods.dev/badge/skills/jackstewart254/groundtruth/tldr.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00094 | $0.00629 |
| Opus 5.5 | $0.00038 | $0.00252 |
| Sonnet 5.5 | $0.00019 | $0.00126 |
| Haiku 4.5 | $0.00009 | $0.00063 |
Grade A, and why
tldr scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 53 lines — stays where its author put it; the contents beside it link to each section on GitHub.
TLDR
The user does not read long status updates. They need exactly three things: is anything broken, what is happening, and what do they have to decide. Everything else is noise to them — however useful it feels to write.
Format
Every status update, progress report, plan summary, or turn-ending message uses this exact template:
**Problems:** none
**Doing:**
- <one line>
- <one line>
**Needs you:** none
- Always all three sections, always in this order — a fixed shape is scannable, a varying one is not.
- 1–3 bullets per section, one line per bullet. No sub-bullets, no paragraph after a bullet.
- A section with nothing in it says
noneon the header line — never omit the header, never pad with filler. - Problems = errors, blockers, failing tests, anything broken or risky. Quote the shortest decisive error line, not the log.
- Doing = what just happened or is in progress, stated as outcomes ("auth middleware fixed") not process narration.
- Needs you = decisions, approvals, or proposals awaiting the user. If a bullet here is a question, make it answerable with one word.
Behavior
- On invocation: if the previous assistant message was a long update, restate it in the template first — that message is why the user invoked this.
- Stays active for the rest of the session. Turn off only on "verbose", "stop tldr", or "normal updates".
- Detail is available on request ("expand problems", "show the log") — compress by default, expand only when asked.
- Applies to status prose only. Content the user explicitly asked for — code, a report, a walkthrough, a document — is delivered in full; the template wraps around it, it does not truncate it.
- Safety-critical text (destructive-action confirmations, security warnings) is written in full clear sentences, then the template resumes.
Example
Instead of:
I've been working through the migration and wanted to give you a detailed update on where things stand. First, I analyzed the existing schema and found that the users table has a legacy column... (12 more paragraphs)
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2mo ago First seen · 53 lines · 94 tokens per session scan A 2e9889368def
tldr is a skill published in the GitHub repository jackstewart254/groundtruth (0 stars, last pushed 3mo ago), licensed MIT. It adds 94 tokens to every session and 629 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-01.
tldr compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| tldrjackstewart254/groundtruth | Skill | 0 | 3mo ago | MIT | A | 94 | Claude Code |
| triage-inboxvscarpenter/gsd-task-manager | Skill | 40 | yesterday | MIT | A | 0 | Claude Code, Codex |
| setup-slack-channelCopilotKit/CopilotKit | Skill | 39k | yesterday | MIT | A | 206 | Claude Code |
| indian-district-hierarchyjayanthmb14/forthepeople | Skill | 313 | 2d ago | MIT | A | 113 | Claude Code, Codex |
| email-formattingprovos/ironcurtain | Skill | 616 | 2d ago | Apache-2.0 | A | 53 | Claude Code, Codex |
| disk-cleanupmtarcure/claude-vibe-squad | Skill | 228 | 9d ago | MIT | A | 92 | Claude Code |
| cortex-profilecdeust/Cortex | Skill | 76 | yesterday | MIT | A | 98 | Claude Code |