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/eugeniughelbur/clearmode/agents-mdgit clone --depth 1 https://github.com/eugeniughelbur/clearmodeWrote 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/eugeniughelbur/clearmode/agents-md)<a href="https://agentmods.dev/instructions/eugeniughelbur/clearmode/agents-md"><img src="https://agentmods.dev/badge/instructions/eugeniughelbur/clearmode/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 | $0.01110 | $0.01110 |
| Opus 5 | $0.00555 | $0.00555 |
| Sonnet 5 | $0.00222 | $0.00222 |
| Haiku 4.5 | $0.00111 | $0.00111 |
Grade A, and why
clearmode 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 4d 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 — 99 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
clearmode is the CLEAR-100 writing standard, 43 rules on 4 axes, plus the checker that scores them. These are the rules for any agent editing this repo.
The one rule that matters
rules/clear-100.json is the single source of truth. Every other rule-bearing file in this repo is generated from it.
Generated, never hand-edited:
SKILL.mdCLEAR-100.mdreferences/rules.mdreferences/lexicons.md- everything under
targets/
To change a rule, a threshold, a word list, or a gloss:
- Edit
rules/clear-100.json. - Run
python3 scripts/compile_targets.py. - Run
python3 -m unittest discover tests -v. - Commit the JSON and the regenerated targets together.
A pull request that edits a generated file by hand gets closed.
Layout
| Path | What it is |
|---|---|
rules/clear-100.json |
every rule, threshold, and lexicon |
rules/preamble.md |
hand-written intro that heads CLEAR-100.md |
scripts/clearcheck.py |
the checker, stdlib only |
scripts/compile_targets.py |
rule pack in, 18 targets out |
commands/ |
slash commands |
tests/ |
unittest suite and fixtures |
install.sh |
detects your agents and installs the right target |
Constraints
- Stdlib only. No pip, no npm, no lockfile. If a feature needs a dependency, the feature is wrong.
- Python 3.10+. Type hints welcome, not required.
- Deterministic. Same input, same score, every run. No model calls in the checker.
- No em-dashes in prose. The standard bans them, so this repo cannot use them.
tests/fixtures/bad.mdis the exception: it exists to contain tells. - No emoji. Anywhere.
- Every rule states its failure mode. A rule with no
whyand nofixdoes not ship.
Adding a rule
- Add the object to
rules.rules[]in the JSON:id,axis,title,severity,detector,weight,why,fix, and abadandgoodpair. - If the detector is
regex, that is all. Otherwise add a_detectormethod onCheckerand dispatch it fromrun(). - Add a fixture case that the rule catches and one it must not.
- Recompile and test.
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.
- 4d ago First seen · 99 lines · 1,110 tokens per session scan A 320a27e6804d
clearmode AGENTS.md is an instructions file published in the GitHub repository eugeniughelbur/clearmode (1 stars, last pushed 5d ago), licensed MIT. It adds 1,110 tokens to every session, about $0.0056 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
shuorenhua AGENTS.md
AGENTS.md instructions for MrGeDiao/shuorenhua, covering 仓库说明, 语言, 风格, 布局 and 协作.
vale-ai-tells CLAUDE.md
Instructions for tbhb/vale-ai-tells, covering claude code instructions, project overview, repository structure, development workflow and rule conventions.
vale-ai-tells AGENTS.md
Instructions for tbhb/vale-ai-tells, covering agent instructions, commit messages, prose lint output and drafting a document.
unmachined AGENTS.md
Instructions for 0xNyk/unmachined, covering agent operating rules, scope, verification, git contract and pull requests.
humanizer AGENTS.md
Instructions for hannsxpeter/humanizer, covering humanizer (agent instructions), when to apply this skill, how to run it, hard rule (faithfulness over liveliness) and scope.
humanizer copilot-instructions.md
Instructions for hannsxpeter/humanizer: This repository is the humanizer skill: pure-prompt instructions that rewrite AI-sounding prose so it reads as genuinely human, and rewrite in a specific writer's voice when a sample or profile is available. It also cleans suspicious invisible Unicode in supplied prose without…