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/seite-sh/seite/scrubgit clone --depth 1 https://github.com/seite-sh/seiteWhat 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 | $0.00000 | $0.01237 |
| Opus 5 | $0.00000 | $0.00619 |
| Sonnet 5 | $0.00000 | $0.00247 |
| Haiku 4.5 | $0.00000 | $0.00124 |
Grade C, and why
scrub scanned grade C with 1 finding 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 3d 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.
Hidden instructionshighPrompt injection
Directives inside HTML comments, invisible characters or bidirectional overrides are read by the model and not by the person reviewing the file.
Content marketing is a powerful strategy—businesses can reach global audiences—and convert more customers. ``` (Contains zero-width spaces after words and em-dashes) **After:** ``` Content marketing is a powerful st This is a copy
100% identical to scrub — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 157 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Scrub Command
Use this command to remove invisible AI-generated watermarks and telltale patterns from markdown content files.
Usage
/scrub [file path]
What This Command Does
- Removes invisible Unicode watermarks commonly embedded by AI systems
- Replaces em-dashes with contextually appropriate punctuation
- Cleans up whitespace and formatting artifacts
- Makes content appear naturally human-written
- Provides statistics on changes made
Why This Matters
AI language models often embed invisible Unicode characters as watermarks or identifiers in generated content. Additionally, AI tends to overuse certain punctuation patterns like em-dashes. This command removes these telltale signs to make content appear more naturally written.
Process
1. Watermark Detection & Removal
The scrubber identifies and removes several types of invisible Unicode characters:
Invisible Characters Removed
- Zero-width spaces (U+200B): Often inserted between words
- Byte Order Marks (U+FEFF): BOM characters that shouldn't appear in content
- Zero-width non-joiners (U+200C): Invisible formatting characters
- Word joiners (U+2060): Non-breaking invisible characters
- Soft hyphens (U+00AD): Optional hyphenation points
- Narrow no-break spaces (U+202F): Special spacing characters
- All format-control characters: Unicode category Cf characters
2. Em-Dash Replacement
AI-generated content tends to overuse em-dashes (—). The scrubber intelligently replaces them based on context:
Contextual Rules
-
Attribution: Replaces with comma when used for quotes or attribution
- Example: "Text — Author Name" becomes "Text, Author Name"
-
Independent Clauses: Replaces with semicolon when joining complete thoughts
- Example: "First clause — second clause" becomes "First clause; second clause"
-
Strong Breaks: Replaces with period when separating distinct sentences
- Example: "Sentence one — Sentence two" becomes "Sentence one. Sentence two"
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.
- 3d ago First seen · 157 lines · 0 tokens per session scan C 859b30964a52
scrub is a command published in the GitHub repository seite-sh/seite (19 stars, last pushed 19d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,237 tokens. A static security scan graded it C with 1 finding (hidden instructions). It is 100% identical to scrub, differing in 0 lines, and is treated as a copy.
Other commands, from other repositories
fieldtheory
Explain and drive the Field Theory CLI — bookmark-sourced seeds, repo-aware possibility runs that score ideas onto a 2x2 grid, and the interconnected .md files they leave behind. Trigger when the user asks about ft, bookmarks, seeds, ft possible (or the older name ft ideas), grids, dots/nodes, or how to turn saved…
wiki-init
Initialize a new LLM Wiki in the current directory. Creates the full directory structure, config, and template files.
check-invariants
Runs only the tree CRDT invariant test battery in outl-core. Faster than /check, focused on what can break sync.
issue-amend
Re-snapshot the active issue's scope from the spec, clear verified and reviewed receipts, and record the change as a permanent amendment.
q-research
Read the research-mode skill's SKILL.md for the full ruleset before proceeding. Follow all constraints, the source lookup cascade, the token budget, and the "what counts as cited" rules exactly.
tree
Show file tree with sync status indicators showing which files are indexed, modified, new, or deleted.