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/datasift-ty-personal/siftstack/claude-mdgit clone --depth 1 https://github.com/DataSift-Ty-Personal/SiftStackWhat 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 | $0.32351 | $0.32351 |
| Opus 5 | $0.16175 | $0.16175 |
| Sonnet 5 | $0.06470 | $0.06470 |
| Haiku 4.5 | $0.03235 | $0.03235 |
Grade A, and why
SiftStack CLAUDE.md scanned grade A with 1 finding 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 2d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- **L3 fallback fetcher (Scrapfly ASP, build 1.0.32+):** `src/scrapfly_browser.py` (`ScrapflyBrowserClient.fetch(url)`, plus a `python src/scrapfly_browser.py <url>` CLI) clears Cloudflare/JS walls on county-record + gen How it starts
The opening of the file, as written. The whole thing — 849 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md — SiftStack
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
Project Overview
SiftStack — Full-stack real estate investing operations platform built around DataSift.ai CRM. Covers the entire REI business lifecycle:
- Data Acquisition: Web scraping tnpublicnotice.com (foreclosures, tax sales, probates), scanned PDF import, courthouse terminal photo import (probate, eviction, code violations, divorce), Dropbox auto-polling
- Enrichment Pipeline: 10+ steps — Smarty address standardization, Zillow property data, Knox County Tax API, obituary/heir research, Ancestry.com SSDI, Tracerfy skip trace, Trestle phone scoring, entity research
- Deal Analysis: Comparable sales (Two-Bucket ARV), rehab estimation (4-tier room-by-room), deal analyzer (MAO/ROI/financing scenarios)
- Market Intelligence: Zip code scoring, Market Finder reports, cash buyer list building, investor portfolio analysis
- CRM Automation: DataSift upload, 26 TCA sequence templates, 12 niche sequential marketing presets, filter preset management, SiftMap sold property tagging
- Lead Management: 4 Pillars of Motivation auto-qualification, STABM daily routine, pipeline reporting, deep prospecting (4-level framework)
- Operations: Acquisition playbook generator (SOPs, scripts, checklists), Slack/Discord notifications, Google Drive upload, Apify Actor deployment
Currently focused on Knox and Blount counties, Tennessee. A realtor sphere-of-influence beta runs on the Columbus OH metro (the soi_* modules; see "Sphere of Influence Pipeline").
- REI Skill Library: 21 Claude Co-Work skill files (
.skill/.pluginZIPs) for distribution to DataSift community via learn.datasift.ai/claude-skills-rei. Skills teach Claude specific REI workflows when uploaded to Co-Work sessions or Projects.
Commands
# Setup
pip install -r requirements.txt
playwright install chromium
cp .env.example .env # then fill in credentials
# Run
python src/main.py daily # new notices since last run
python src/main.py historical # last 12 months of data
python src/main.py daily --split # separate CSV per county+type
python src/main.py daily --counties Knox # only Knox county
python src/main.py daily --types foreclosure,probate # only specific types
python src/main.py daily -v # verbose/debug logging
# Comp package (boundary-filtered comps + dual-track ARV + rehab + buyers -> Excel)
python src/comp_package.py --address "158 Old State Rd" --zip 37914 \
--beds 2 --baths 1 --sqft 1946 --year-built 1938 \
--bbox "35.996,36.016,-83.895,-83.840" --streets "old state|nash rd|seahorn"
# Post-walkthrough package (comps + rehab matrix + walk findings + exits + dispo, Sift-linked)
python src/post_walkthrough.py --walkthrough-template # writes walkthrough_template.json
python src/post_walkthrough.py --address "158 Old State Rd" --city Knoxville --zip 37914 \
--bbox "35.996,36.016,-83.895,-83.840" --streets "old state|nash rd|seahorn" \
--walkthrough walk_158.json --buyers output/buyer_sweep_37914_20260723.json \
--outreach output/dispo_skiptrace_158.json
python src/post_walkthrough.py --address "..." --sold-json output/zillow_37914_sold.json # free re-run
# DataSift preset/sequence management
python src/main.py manage-presets --discover # list all presets and sequences
python src/main.py manage-presets --add-sold-exclusion # add Sold exclusion to all presets
python src/main.py manage-presets --create-sold-sequence # create Sold cleanup sequence
python src/main.py manage-presets --all # discovery + update + sequence
# SiftMap sold property tagging
python src/main.py manage-sold --months-back 12 # tag sold properties (last 12 months)
python src/main.py manage-sold --counties Knox --min-sale-price 5000
# Courthouse photo import (build 1.0.28+)
python src/main.py photo-import --folder ./photos --photo-county Knox --photo-type probate
python src/main.py photo-import --folder ./photos --photo-county Knox --photo-type eviction --skip-obituary
python src/main.py dropbox-watch # auto-poll Dropbox for new photos
python src/main.py dropbox-watch --poll-interval 300 --max-polls 5 # 5-min interval, 5 cycles
python src/main.py dropbox-watch --no-delete # keep photos in Dropbox after processing
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.
- 2d ago First seen · 849 lines · 32,351 tokens per session scan A 2be6a80d1096
SiftStack CLAUDE.md is an instructions file published in the GitHub repository DataSift-Ty-Personal/SiftStack (21 stars, last pushed 5d ago), licensed MIT. It adds 32,351 tokens to every session, about $0.1618 per session on Opus 5. A static security scan graded it A with 1 finding (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
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.
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).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.