Borrowing it
Nothing to install: this file belongs to froggeric/gemini-watermark-and-synthid-remover. 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/froggeric/gemini-watermark-and-synthid-remover/main/CLAUDE.mdgit clone --depth 1 https://github.com/froggeric/gemini-watermark-and-synthid-removerWrote 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/froggeric/gemini-watermark-and-synthid-remover/claude-md)<a href="https://agentmods.dev/instructions/froggeric/gemini-watermark-and-synthid-remover/claude-md"><img src="https://agentmods.dev/badge/instructions/froggeric/gemini-watermark-and-synthid-remover/claude-md/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/instructions/froggeric/gemini-watermark-and-synthid-remover/claude-md"><img src="https://agentmods.dev/badge/instructions/froggeric/gemini-watermark-and-synthid-remover/claude-md.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 | $0.20356 | $0.20356 |
| Opus 5 | $0.10178 | $0.10178 |
| Sonnet 5 | $0.04071 | $0.04071 |
| Haiku 4.5 | $0.02036 | $0.02036 |
Grade C, and why
gemini-watermark-and-synthid-remover CLAUDE.md scanned grade C with 2 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 7d 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
- **HF model uploads (`froggeric/wmr`):** use `hf upload froggeric/wmr <local> <path-in-repo>` (NOT `huggingface-cli`, which is deprecated and no-ops); it auto-uses the cached `~/.cache/huggingface/token` and auto-tracks Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- **Download progress + leftover cleanup (1.16.11):** an HF fetch is 302 -> CDN and curl fires the xferinfo cb during the redirect body too, reporting the ~1 KB redirect page's Content-Length as the total; `download_pinn How it starts
The opening of the file, as written. The whole thing — 421 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
Build Commands
Recommended (macOS/Homebrew), resilient to brew upgrade:
scripts/build.sh # Release build + tests; self-heals a stale cache
BUILD_TYPE=Debug scripts/build.sh
RUN_TESTS=0 scripts/build.sh # build only
scripts/build.sh verifies the required Homebrew formulas, auto-wipes + reconfigures when a cached dependency path has vanished after an upgrade, and configures against the stable /opt/homebrew/opt/… symlinks. Binaries: build/wmr, build/tests/wmr_tests. Test suite needs Catch2 (brew install catch2).
Or, arm64 preset: cmake --preset mac-homebrew-Release && cmake --build --preset mac-homebrew-Release.
System libs (macOS/Homebrew), manual, no vcpkg:
cmake -B build -S . -GNinja \
-DCMAKE_PREFIX_PATH="$(brew --prefix opencv);$(brew --prefix ffmpeg);$(brew --prefix catch2);$(brew --prefix fmt);$(brew --prefix spdlog);$(brew --prefix cli11)" \
-DOpenCV_DIR=$(brew --prefix opencv)/lib/cmake/opencv4 \
-DFFMPEG_ROOT=$(brew --prefix ffmpeg) \
-DWMR_BUILD_TESTS=ON
cmake --build build
vcpkg (all platforms):
cmake -B build -S . -GNinja
cmake --build build
Tests:
ctest --test-dir build --output-on-failure
./build/tests/wmr_tests "[v2]" # single tag (path: tests/wmr_tests)
Integration tests need project root as CWD (they look for test-images/ relative to CWD). Tests use SKIP macro if test data is absent, so they don't fail without it.
Architecture
Single-pass C++20 CLI tool. No libraries, everything compiles into one wmr executable.
Pipeline: Detect → Remove → Inpaint
WatermarkEngine (src/core/) orchestrates the image pipeline:
- NccDetector (detection/), 3-stage NCC: spatial template match (cv::matchTemplate), gradient match (Sobel magnitudes), variance analysis. Fusion: spatial×0.50 + gradient×0.30 + variance×0.20. Threshold: 0.35.
- Reverse alpha blend (core/blend_modes),
original = (watermarked - alpha*255) / (1-alpha). Alpha maps decoded from embedded PNGs (assets/embedded_assets.hpp). - Inpaint (core/inpaint) — OFF by default since 1.14.0: the default is the pure
reverse-alpha-blend (the exact inversion). The opt-in
--denoise ns|teleacleanup is residual-only (predicts clean content with cv::inpaint and blends it in only where the reverse-blend left a real residual, never blurring a clean removal).
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.
- 7d ago Changed · +8 lines · +822 tokens per session a04a29dfcbc5
- 11d ago First seen · 413 lines · 19,534 tokens per session scan C 007283a21a2c
gemini-watermark-and-synthid-remover CLAUDE.md is an instructions file published in the GitHub repository froggeric/gemini-watermark-and-synthid-remover (63 stars, last pushed 7d ago), licensed MIT. It adds 20,356 tokens to every session, about $0.1018 per session on Opus 5. A static security scan graded it C with 2 findings (recursive force delete, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).