Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/pehcastro/harness/tldrgit clone --depth 1 https://github.com/pehcastro/harnessWrote 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/commands/pehcastro/harness/tldr)<a href="https://agentmods.dev/commands/pehcastro/harness/tldr"><img src="https://agentmods.dev/badge/commands/pehcastro/harness/tldr.svg" alt="Measured on agentmods" 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 | $0.00029 | $0.00537 |
| Opus 5 | $0.00015 | $0.00269 |
| Sonnet 5 | $0.00006 | $0.00107 |
| Haiku 4.5 | $0.00003 | $0.00054 |
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 5d 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.
What it actually says
tldr
Compress recent output into a TLDR. The reader has already read it and wants the short version, so the summary is the whole reply.
What to summarise
Read $ARGUMENTS.
| Argument | Summarise |
|---|---|
| empty | Your last reply |
session |
Everything since the session started |
last N |
Your last N replies |
| anything else | Treat it as the subject and summarise what was said about it |
Shape
Answer with the summary alone. No heading, no preamble, no offer to expand. The
reader asked for the short version, so a TLDR label on a reply that is entirely
a TLDR is noise.
Caps: 40 words for the last reply, 150 for a session. Both are hard.
What survives compression
Keep, in this order of priority:
- The decision the reader has to make, and the options.
- Anything still open or unanswered.
- A risk, and anything that cannot be undone.
- Numbers the reader needs: a count, a file path, an identifier.
- What changed, when the reader cannot see it.
Drop everything else. Reasoning, evidence, how you got there, and what you tried first are all gone. If it survives the cut it was not decoration.
Rules that still apply
Every rule in the output style holds here, since this is chat output like any other. No banned words, no em dash, no preamble, no closing offer.
One that matters more than usual: do not restate what the reader watched. A session TLDR is not a list of every command you ran. It is where things stand now.
Examples
A long reply about a failing lookup:
Not the lookup. No user with id `A` is seeded, so /users/A 404s correctly.
Decide: seed a user `A`, or make the lookup case-insensitive.
A session that built an API:
Hono API in `src/`, 5 tests passing. Store is an in-memory Map, so it
resets on restart. Two open: whether the 404 body should include the id,
and whether to persist before the demo.
When the reader wants more
They will ask. Do not offer.
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.
- 5d ago First seen · 72 lines · 29 tokens per session scan A 2f8881918cc8
tldr is a command published in the GitHub repository pehcastro/harness (1 stars, last pushed 18d ago), licensed MIT. It adds 29 tokens to every session and 537 once invoked, about $0.0001 per session on Opus 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 2026-08-31.
Other commands, from other repositories
setup-agents
Scaffold a multi-agent AI setup for any repo. Scans the codebase, interviews the developer, generates agent configs for whichever AI tools the repo uses (Copilot, Claude Code, Cursor, Codex, Windsurf). Use when asked to "set up agents", "scaffold Copilot agents", or "create an AGENTS.md".
lessons-compact
Compact MEMORY/LESSON.md by deduplicating near-identical lessons, merging same-root-cause lessons, and pruning stale ones — without losing any unique signal. Delegates the analysis to the lessons-compactor agent.
prompt-history
Manage history of created and optimized prompts.
lessons
View project lessons in MEMORY/LESSON.md (project root) and append a compact "never reproduce" lesson.
download
Download Modellix task results to a local directory before the resource URLs expire.
tasks
Inspect Modellix task status and recover tasks after a timeout or unknown submission.