Borrowing it
Nothing to install: this file belongs to hannsxpeter/authenticity-check. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/hannsxpeter/authenticity-check/main/.windsurfrulesgit clone --depth 1 https://github.com/hannsxpeter/authenticity-checkWrote 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/rules/hannsxpeter/authenticity-check/windsurfrules)<a href="https://agentmods.dev/rules/hannsxpeter/authenticity-check/windsurfrules"><img src="https://agentmods.dev/badge/rules/hannsxpeter/authenticity-check/windsurfrules/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/rules/hannsxpeter/authenticity-check/windsurfrules"><img src="https://agentmods.dev/badge/rules/hannsxpeter/authenticity-check/windsurfrules.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00627 | $0.00627 |
| Opus 5.5 | $0.00251 | $0.00251 |
| Sonnet 5.5 | $0.00125 | $0.00125 |
| Haiku 4.5 | $0.00063 | $0.00063 |
Grade A, and why
windsurfrules 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 19d 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
authenticity-check (Windsurf rules)
This repository is the authenticity-check skill: pure-prompt instructions
that score how authentically a piece of text reads as the work of a real
human author and flag the spans that read as AI-generated, AI-templated, or
generically derivative, plus suspicious Unicode provenance carriers in the
supplied text. No scripts, no dependencies, no network access; tools are
read-only. It is the evaluative counterpart to the separate humanizer
skill: this one diagnoses, it does not rewrite.
Apply this skill when the user wants to know whether a text is authentic, whether it reads like an LLM or like a person, how human a passage sounds, which parts sound machine-written, whether a draft still sounds like the author or a named writer, or whether pasted text contains a hidden AI watermark or invisible Unicode. Apply it even when the words "authenticity check" are not used and even when the cue is oblique ("does this sound like a bot," "this feels generated"). Do not apply it to transformative requests (humanize, de-slop, de-AI, rewrite, fix, remove marks, make it sound like X); those belong to a separate transformation or provenance hygiene skill.
Read SKILL.md at the repository root and follow it exactly, with
references/tell-patterns.md (32 patterns, six families),
references/do-not-flag.md, references/provenance-signals.md when needed,
and, in voice-deviation mode only,
references/voice-matching.md. Run the method as written: Step 0 baseline
discovery, Step 0a provenance preflight, Step 0b density pre-check, then the
multi-pass diagnostic (catalog
scan, mandatory false-positive audit with veto power, read-only
internal-consistency heuristics, voice deviation if a voice target exists).
Emit the exact output contract: Authenticity report (band plus 0-100 score) /
Provenance signals / Flagged spans / Reads as human / Score basis / Caveats /
Next step.
Hard rule: this skill scores and flags only. It never rewrites, edits, removes
or normalizes Unicode, strips metadata, or returns improved prose, not even
one suggested span; the rewrite is the
separate humanizer skill's job, run as a human-judged step, with no target
score carried into it. A combined score-then-rewrite loop is detector-gaming,
which humanizer refuses. This skill is not for defeating AI-detection
systems; reframe such requests toward an honest read of quality and voice.
references/tell-patterns.md, references/do-not-flag.md, and
references/voice-matching.md are synced copies vendored from the canonical
humanizer repo. Do not edit them here; re-sync from humanizer when its
criteria change (see each file's header stamp and the README).
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.
- 19d ago First seen · 46 lines · 627 tokens per session scan A 9f35d918beb6
windsurfrules is a cursor rule published in the GitHub repository hannsxpeter/authenticity-check (2 stars, last pushed 19d ago), licensed MIT. It adds 627 tokens to every session, about $0.0025 per session on Opus 5.5. 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 2026-10-08.
Other cursor rules, from other repositories
humanizer
Humanize AI-sounding prose, rewrite in a writer's voice, and conservatively clean suspicious invisible Unicode in supplied prose. Use whenever the user wants to humanize, de-slop, de-AI, or de-robotify text; fix writing that sounds like an LLM, corporate, salesy, generic, or off; make a draft sound like them or a…
patina
Load Patina's public repository development rules.
cursor-rules
Anti-AI-tell writing rules. Apply to all prose output (docs, comments, commit messages, README content, user-facing strings).
anti-ai-writing-humanizer
Humanize prose — write or rewrite so it doesn't read as AI. Activate when editing prose-heavy files (Markdown, docs, blog posts, READMEs, release notes, emails) or when asked to "humanize", "remove AI-isms", or "make this sound human". Cursor port of the anti-ai-writing-humanizer skill. See…
clearmode
CLEAR-100 writing standard. Plain, human, structured, dense.
caveman-stats
This skill is delivered by hooks/caveman-stats.js (read by hooks/caveman-mode-tracker.js on /caveman-stats). The model does not need to do anything when this skill fires — the hook returns decision: "block" with the formatted stats as the reason. The user sees the numbers immediately.