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/wawimundo/viber-mcp/claude-mdgit clone --depth 1 https://github.com/wawimundo/viber-mcpWrote 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/wawimundo/viber-mcp/claude-md)<a href="https://agentmods.dev/instructions/wawimundo/viber-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/wawimundo/viber-mcp/claude-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 5.1 | $0.02068 | $0.02068 |
| Opus 5 | $0.01034 | $0.01034 |
| Sonnet 5 | $0.00414 | $0.00414 |
| Haiku 4.5 | $0.00207 | $0.00207 |
Grade A, and why
viber-mcp CLAUDE.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 — 133 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.
What this is
A macOS-only MCP server that sends and reads Viber Desktop messages by automating the running,
signed-in Viber app — Viber has no API, its message DB is encrypted, and its Qt accessibility tree
is unusable, so everything is done via the viber:// URL scheme, synthetic HID input, and screen OCR.
Commands
uv sync # install deps (Python managed by uv)
# Standalone client = the way to test/run tools without Claude (spawns a fresh server each call,
# so it always uses the latest code). Subcommands map 1:1 to tools:
uv run client.py list # advertise tools
uv run client.py open|openread|search "<name-or-+E164>"
uv run client.py send "<to>" "<text>" [--dry-run]
uv run client.py reply "<text>" # into the currently-open chat
uv run client.py draft "<text>" # paste into open chat, do NOT send
uv run client.py note "<text>" # save to My Notes
uv run client.py chats | read | contacts
# Register with Claude Code (user scope):
claude mcp add --transport stdio --scope user viber -- uv --directory "$(pwd)" run server.py
There is no test suite, linter, or build step. Verify changes by driving the real app with
client.py and screenshotting the Viber window (viber_ocr.capture_window) to confirm the outcome —
the returned verified flag is best-effort and gives false negatives.
After editing server code, the registered MCP server keeps running the old code until reconnected
(/mcp → reconnect viber, or restart Claude Code); client.py always runs fresh code.
Architecture
The code is split into platform-neutral domain logic and a platform Backend (so the app is port-ready for Windows/Linux — see "Porting" below):
server.py— FastMCP stdio server; thin@mcp.tool()wrappers only. Tools:send_message,reply,open_chat,open_and_read,read_open_chat,list_chats,put_draft,search_and_open,save_to_my_notes,open_my_notes,list_contacts.viber_send.py— platform-neutral write/navigation logic: recipient resolution (contacts.jsonalias → E.164), the send pipeline, search-result scoring, panel detection, send verification. Holds a module-levelbackend = get_backend()and calls only its semantic methods. RaisesSendError.viber_ocr.py— platform-neutral read side:read_open_chat()/read_chat_list()formatbackend.observe()output (read_chat_listkeeps only left-column x<0.30). RaisesOCRError.backends/— the platform layer.base.py—BackendABC +OCRError. Semantic interface:app_running/focus/open_url,observe()→[{text,x,y}], clipboard,type_text/select_all/delete/paste, and element clicksclick_compose/send/search/result(norm_y)/panel_close. The domain never sees pixels or key codes.macos.py—MacBackend: all Quartz/Vision/AppKit code AND the macOS Viber-layout constants (element pixel/fraction offsets). This is the only file with OS APIs.__init__.py—get_backend()picks the backend bysys.platform(macOS only today).
client.py— standalone MCP stdio client for testing.contacts.json— name→E.164 map, git-ignored (real numbers); shipcontacts.example.json.
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 · 133 lines · 2,068 tokens per session scan A 1fd0bfe52bbe
viber-mcp CLAUDE.md is an instructions file published in the GitHub repository wawimundo/viber-mcp (0 stars, last pushed 1mo ago), licensed MIT. It adds 2,068 tokens to every session, about $0.0103 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 instructions, from other repositories
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.
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.
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).