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.
git 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/commands/eugeniughelbur/clearmode/clear-rewrite)<a href="https://agentmods.dev/commands/eugeniughelbur/clearmode/clear-rewrite"><img src="https://agentmods.dev/badge/commands/eugeniughelbur/clearmode/clear-rewrite.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 | $0.00040 | $0.00470 |
| Opus 5 | $0.00020 | $0.00235 |
| Sonnet 5 | $0.00008 | $0.00094 |
| Haiku 4.5 | $0.00004 | $0.00047 |
Grade A, and why
clear-rewrite 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 6d 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
/clear-rewrite
Rewrite until it passes. Show the work.
Steps
-
Run
/clear-checkfirst. You need the baseline score before you touch a word. -
Fix in this order. The order matters, because each pass makes the next one cheaper.
- Cut. Delete every sentence that fails the delete test (D1, D3, D4). Do this before rewriting anything. Half the findings disappear here.
- Unslop. Replace every H-axis hit with the specific thing it stood in for. If nothing specific is behind it, the sentence was already deleted in step 1.
- Plain. Split long sentences. Swap long words for short ones. Gloss each term of art once. Put the actor before the verb.
- Structure. Turn parallel items into lists, sequences into numbered steps, comparisons into tables. Break walls.
- Rhythm. Read it out loud. Put a short sentence next to a long one. This is the last pass, never the first.
-
Re-run the checker. Repeat from the highest remaining severity. Stop at the gate, or after four passes.
-
If four passes do not clear the gate, say so and name what is blocking. Do not fake the score.
Output
- The rewritten text in full, in a fenced block, ready to paste.
- One line:
before X/100 -> after Y/100, with the axis that moved most. - A short list of what you cut and why. Cuts are the part a user will want to argue with.
Hard limits
- Never change a claim, a number, a name, or a link. Clarity work only. If a claim is wrong, say so separately.
- Never touch code, code blocks, quoted material, or another person's words.
- Never remove a technical term that is the subject. Gloss it once instead.
- Never add a claim the source did not make to satisfy the density axis.
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.
- 6d ago First seen · 35 lines · 40 tokens per session scan A 4d41c7499f7a
clear-rewrite is a command published in the GitHub repository eugeniughelbur/clearmode (1 stars, last pushed 8d ago), licensed MIT. It adds 40 tokens to every session and 470 once invoked, about $0.0002 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
write-post
Draft a LinkedIn text post in the user's voice with inline humanizer drafting constraints.
analyze-performance
Diagnose what is landing across the user's recent LinkedIn posts.
audit-profile
Diagnose the user's LinkedIn profile section by section with severity-rated fixes.
build-carousel
Produce a LinkedIn carousel SPEC (slide copy + visual brief + image-gen prompts) for assembly in Canva or Figma.
find-ideas
Interview the user to surface specific, postable LinkedIn ideas from their actual work.
plan-content
Define LinkedIn content pillars, audience, and posting cadence.