social_dive CLAUDE.md

social_dive CLAUDE.md is an instructions file for coding agents from krishddd/social_dive. It costs 521 tokens per session, scanned A, original, MIT.

A project instruction file describing development rules for Social Dive, a Python and Rust tool for connecting agents to social-media channels.

In plain words
What is it for?
Use it to guide changes in the Social Dive project, including channel implementations, credential handling, version synchronization, and test expectations.
Why use it?
It gives agents consistent rules for code style, architecture, configuration, logging, version updates, and testing.

Instructions file

Install

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.

agentmods
npx agentmods add instructions/krishddd/social_dive/claude-md
Clone the repo
git clone --depth 1 https://github.com/krishddd/social_dive

Wrote 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.

agentmods badge for social_dive CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/krishddd/social_dive/claude-md.svg)](https://agentmods.dev/instructions/krishddd/social_dive/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/krishddd/social_dive/claude-md"><img src="https://agentmods.dev/badge/instructions/krishddd/social_dive/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 521 This file is loaded in full into every session.
When invoked 521 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00521 $0.00521
Opus 5 $0.00260 $0.00260
Sonnet 5 $0.00104 $0.00104
Haiku 4.5 $0.00052 $0.00052

Measured 3d ago against content hash 396c692c68dd, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

social_dive 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 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.

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.

CLAUDE.md · 46 lines

How it starts

The opening of the file, as written. The whole thing — 46 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Social Dive — Developer Conventions

Language & Environment

  • Python 3.10+ with type hints on every function/method signature
  • Rust (edition 2021) for the _social_dive_core extension module via PyO3/maturin
  • loguru for all logging (never print() for diagnostic output)
  • rich for all CLI terminal output
  • httpx for Python-side HTTP requests (async-capable)
  • openai SDK for all OpenAI-compatible API calls (NVIDIA NIM, OpenAI, MiniMax)

Architecture Rules

  • This is a glue/routing layer, not a wrapper — agents call upstream tools directly after setup
  • Never modify upstream open-source tool source code
  • Every channel must implement all four abstract methods: can_handle(), read(), search(), check()
  • A broken channel must never crash the doctor report or affect other channels
  • Config/credentials stored only at ~/.social-dive/config.yaml with 0600 permissions
  • Environment variables always override config file values

Version Sync

Version string must stay in sync across:

  1. pyproject.toml[project].version
  2. social_dive/__init__.py__version__
  3. Cargo.toml[package].version
  4. tests/test_cli.py → version assertion

Testing

  • pytest tests/ -v must pass before any commit
  • Every channel needs: URL pattern test, mock-response parse test, error-handling test
  • test_channel_contracts.py uses reflection to verify all Channel subclasses implement the interface
  • Work on a branch, PR to main

Code Style

  • ruff for linting (pyproject.toml has the config)
  • mypy for type checking
  • Max line length: 100 characters
  • Use from __future__ import annotations in every module
  • Docstrings: Google style

Adding a New Channel

  1. Create social_dive/channels/<name>.py
  2. Subclass Channel, set name, tier, backends
  3. Implement can_handle(), read(), search(), check()
  4. Decorate the class with @register_channel
  5. Add tests in tests/test_channels.py
  6. Auto-discovery handles the rest — no manual registration needed

Read the full file on GitHub · 46 lines

Changes

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.

  1. 3d ago First seen · 46 lines · 521 tokens per session scan A 396c692c68dd

Subscribe to this mod's changes

social_dive CLAUDE.md is an instructions file published in the GitHub repository krishddd/social_dive (1 stars, last pushed 2mo ago), licensed MIT. It adds 521 tokens to every session, about $0.0026 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.