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 instructions/conorbronsdon/grokkable-output/agents-mdgit clone --depth 1 https://github.com/conorbronsdon/grokkable-outputWrote 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/instructions/conorbronsdon/grokkable-output/agents-md)<a href="https://agentmods.dev/instructions/conorbronsdon/grokkable-output/agents-md"><img src="https://agentmods.dev/badge/instructions/conorbronsdon/grokkable-output/agents-md.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 | $0.00509 | $0.00509 |
| Opus 5 | $0.00254 | $0.00254 |
| Sonnet 5 | $0.00102 | $0.00102 |
| Haiku 4.5 | $0.00051 | $0.00051 |
Grade A, and why
grokkable-output AGENTS.md 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.
How it starts
The opening of the file, as written. The whole thing — 37 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent instructions — grokkable-output
This repo ships one artifact: SKILL.md, a portable agent skill that changes how
an agent writes replies, reports, and status updates (verdict first, one-pass
readable, calibrated claims). Everything else supports it.
Layout
SKILL.md— the skill. The only file consumers need.examples/— real, unedited before/after pairs from A/B testing on Opus subagents. Generated artifacts: regenerate only by re-running the eval harness in the development repo, never by hand-editing (they are evidence, not prose).evals/— the corpus behind the testing claims: fixtures, assertion sets, per-assertion graded runs (including the rejected v0.4/v0.5 rounds), analyst summary. Provenance and known artifact defects:evals/README.md.README.md— positioning, install, testing summary, related work.
Rules for working in this repo
- Apply the skill to your own output here. A PR description or issue reply in this repo that buries its lede is a bug report against the product. Unenforced — no hook checks this; reviewers do.
- Examples and eval runs are evidence. Never edit files under
examples/orevals/runs/to read better. Their value is that they are verbatim model output — defects included (seeevals/README.mdfor the known ones). Unenforced by tooling; provenance is documented inexamples/README.mdandevals/README.md. - Testing claims in README.md must trace to the eval corpus. Scores (21/22,
15/19, word counts) come from the graded runs in
evals/runs/and the summary inevals/RESULTS.md. If you change a claim, cite the grading it comes from. Unenforced — no CI yet; treat any unsourced number as a defect (the skill's own rule). - Keep the SKILL.md frontmatter complete (name, description, version, license, compatibility, metadata) — agentskills.io and marketplace listers read it.
- Version bumps follow semver on skill content: pattern/rule changes are minor, wording fixes are patch.
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 · 37 lines · 509 tokens per session scan A f039611dc94d
grokkable-output AGENTS.md is an instructions file published in the GitHub repository conorbronsdon/grokkable-output (2 stars, last pushed 7d ago), licensed MIT. It adds 509 tokens to every session, about $0.0025 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 instructions, from other repositories
avoid-ai-writing CLAUDE.md
Claude Code instructions for conorbronsdon/avoid-ai-writing, covering claude.md, what this is, repository structure, how to make changes and architecture of the skill.
SkillFlow AGENTS.md
AGENTS.md instructions for shinerio/SkillFlow, covering agents.md, directory organization rule - mandatory, documentation organization rule - mandatory, documentation sync rule - mandatory and configuration documentation sync rule - mandatory.
creator.skill AGENTS.md
Instructions for arctan303/creator.skill, covering identity, goals, decision flow, 1. 建立事实 and 2. 选择路线.
unmachined AGENTS.md
Instructions for 0xNyk/unmachined, covering agent operating rules, scope, verification, git contract and pull requests.
datoon CLAUDE.md
Claude Code instructions for andrii-su/datoon, covering claude.md - datoon, project overview, repository layout, source of truth and mirrored files.
token-the-ripper GEMINI.md
Instructions for M4NT/token-the-ripper, a project described as: A skill for your coding agent. Strips scaffolding, keeps signal. Action first. No filler. No hallucination.