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/numen-tech/slopornot/agents-mdgit clone --depth 1 https://github.com/numen-tech/slopornotWrote 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/numen-tech/slopornot/agents-md)<a href="https://agentmods.dev/instructions/numen-tech/slopornot/agents-md"><img src="https://agentmods.dev/badge/instructions/numen-tech/slopornot/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 | $0.03682 | $0.03682 |
| Opus 5 | $0.01841 | $0.01841 |
| Sonnet 5 | $0.00736 | $0.00736 |
| Haiku 4.5 | $0.00368 | $0.00368 |
Grade A, and why
slopornot 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 — 154 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SlopOrNot: agent guide
Brief for AI coding agents (Claude Code, Codex, Cursor, Gemini CLI, Aider) editing this repo. The runtime skill itself is skills/agentic-humanizer/SKILL.md; this file is for agents working on the repo, not running the skill.
What this repo is
SlopOrNot is a plugin bundle for assistant workflows built around Slop or Not.
It ships two skills. agentic-humanizer runs a full 5-pass humanization
workflow with saved preferences and optional voice matching. It works without
Slop or Not; Slop or Not Pro only adds on-device AI detector scoring,
readability checks, Text Cleanup before and after humanization, and cleanup
stats. slop-check is a
self-contained one-shot router for the same on-device tools (text and image
detection, readability, cleanup, status) with no interview and no harness
routing files.
Layout
| Path | Role |
|---|---|
skills/agentic-humanizer/SKILL.md |
Self-contained agentic-humanizer orchestrator. Steps 1-7 (harness detect, profile commands, preferences, voice, optional Slop probe, loop, output). |
skills/agentic-humanizer/harnesses/{claude-code,codex,cursor,gemini-cli,opencode,generic}.md |
Per-harness interview protocols. Edit only the file for the harness you're targeting. |
skills/agentic-humanizer/references/patterns.md |
33-pattern rewrite vocabulary. |
skills/agentic-humanizer/references/detection-guidance.md |
English-only false-positive guard: what not to flag, and human-writing signals to preserve. Loaded on English runs alongside patterns.md. |
skills/agentic-humanizer/references/supplemental-ai-tells.md |
SlopOrNot-authored supplemental AI-tell checks inspired by Wikipedia's field guide. Language-agnostic S1 to S8 concepts, loaded on every run. |
skills/agentic-humanizer/references/multilingual.md |
Multilingual readability registry: supported languages, BCP-47 variants, readability formula per language, reading-level band mapping, code normalization (Norwegian Bokmal to nb). Single source of truth for non-English runs. |
skills/agentic-humanizer/references/profile-resolution.md |
Decision table for SKILL.md Step 3 rule 3: how a saved profile resolves against an unambiguously detected different language (language, variant, reading level, tone, length, English target_grade). Loaded only on the saved-profile-versus-detected-language path. |
skills/agentic-humanizer/references/ai-tells/<code>.md |
Per-language AI-tell catalogues (es, de, it, sv, da, no). Loaded when the resolved language is not English. no.md covers both Bokmal and Nynorsk. |
skills/agentic-humanizer/references/per-iteration-strategies.md |
The 5-iteration cookbook for Core mode and Slop or Not Pro, plus mid-flight Pro-gate fallback. |
skills/agentic-humanizer/references/voice-fingerprint.md |
Voice sample policy, fingerprint schema, extraction prompt, cache rules, and loop injection contracts. |
skills/agentic-humanizer/references/slop-{cli,mcp}-setup.md |
User-facing install guides. |
skills/agentic-humanizer/examples/sample-ai-text.md |
Smoke-test fixture (English). |
skills/agentic-humanizer/examples/sample-ai-text-de.md |
German smoke-test fixture (non-English path). |
skills/agentic-humanizer/README.md |
Dedicated Agentic Humanizer README for users and search indexing. |
skills/slop-check/SKILL.md |
Self-contained slop-check orchestrator. Steps 1-5 (identify op, resolve backend MCP/CLI/app-bundle fallback, run, format, fallback). |
skills/slop-check/references/slop-tools.md |
Full CLI + MCP tool surface for slop-check: params, flags, JSON field paths, score normalization, Pro-gating. |
skills/slop-check/references/slop-setup.md |
slop-check install, Pro unlock, app-bundle fallback, MCP/CLI registration. |
skills/slop-check/README.md |
Dedicated Slop Check README for users and search indexing. |
claude-skills/agentic-humanizer/SKILL.md |
Hand-authored Claude Desktop variant (no harness routing; built-in ask_user_input_v0 interview). Canonical source for the Desktop build. |
claude-skills/agentic-humanizer/README.md |
Hand-authored Claude Desktop install/use guide. |
claude-skills/agentic-humanizer/{references,examples}/ |
Copied verbatim from skills/agentic-humanizer/ by make -C claude-skills build. Do not hand-edit. |
claude-skills/Makefile |
Builds and zips the Claude Desktop bundle (build, zip, check, clean). |
plugins/codex/slopornot/ |
Generated Codex plugin payload. Do not edit synced skill files here by hand. |
plugins/claude/slopornot/ |
Generated Claude Code plugin payload. Do not edit synced skill files here by hand. |
.agents/plugins/marketplace.json |
Codex repo marketplace for the slopornot plugin. |
.claude-plugin/marketplace.json |
Claude Code marketplace for the slopornot plugin. |
scripts/check-{frontmatter,links}.mjs |
Lint scripts run by CI. |
scripts/sync-plugins.mjs |
Copies canonical runtime files into plugin payloads, with --check drift detection. |
scripts/check-plugin-packaging.mjs |
Validates plugin manifests, marketplaces, required files, and sync state. |
scripts/check-versions.mjs |
Verifies the six source version fields agree. CI gate. |
scripts/version-fields.mjs |
Shared source-version-field list and read/write helpers for check-versions.mjs and prepare-release.mjs. |
scripts/prepare-release.mjs |
Release-prep transform: promotes [Unreleased], bumps all version fields, regenerates payloads. |
scripts/draft-highlights.mjs |
Optional DeepSeek-drafted release-notes Highlights, injected atop the promoted CHANGELOG section during prepare. Non-blocking: no-op without DEEPSEEK_API_KEY. |
Makefile |
make dist builds both release zips: claude-skills/agentic-humanizer-claude-desktop.zip and agentic-humanizer-chatgpt.zip. |
.github/workflows/release-{prepare,publish}.yml |
Two-phase release automation: dispatch opens a release PR (with optional DeepSeek highlights); merge tags and uploads the Desktop and ChatGPT zips. |
.github/workflows/prompt-review.yml |
Advisory prompt-quality review on prompt-surface PRs. Not a required check. |
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 · 154 lines · 3,682 tokens per session scan A 3b2a73d9840b
slopornot AGENTS.md is an instructions file published in the GitHub repository numen-tech/slopornot (45 stars, last pushed 2mo ago), licensed MIT. It adds 3,682 tokens to every session, about $0.0184 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 2026-08-30.
Other instructions, from other repositories
im-not-ai CLAUDE.md
Instructions for epoko77-ai/im-not-ai, covering humanize kr — ai 한글 티 제거 하네스 (v2.3.2), 프로젝트 개요, 철칙, ai 워터마킹에 대한 입장 (2026-08-17 확정) and 디렉토리 구조.
im-not-ai GEMINI.md
Instructions for epoko77-ai/im-not-ai, covering humanize kr — ai 한글 티 제거 (gemini cli extension), 개요, 커스텀 명령, 철칙 (위반 시 즉시 롤백) and do-not (탐지·윤문 모두 제외).
humanizer-skill AGENTS.md
AGENTS.md instructions for Aboudjem/humanizer-skill, covering agents.md — humanizer skill, what this repo is, file map, how to install the skill and how to invoke the skill.
unslop CLAUDE.md
Claude Code instructions for MohamedAbdallah-14/unslop, covering unslop — maintainer guide, readme is a product artifact, what this repo is, source of truth and mirrored locations (do not edit by hand).
humanizer-ru AGENTS.md
AGENTS.md instructions for Vladimir-Human/humanizer-ru, covering agents.md — карта репозитория для агентов-редакторов, что это за репозиторий, инварианты (нарушение ломает ci), полный чек-лист одной командой and куда что класть.
vale-ai-tells CLAUDE.md
Instructions for tbhb/vale-ai-tells, covering claude code instructions, project overview, repository structure, development workflow and rule conventions.