Precursor Trends 2025: Signals from Darknet Marketplaces (Q3 Snapshot)

Abstract. This report analyzes precursor-related listings on darknet marketplaces for Q3 2025 using an evidence-backed OSINT dataset (2,565 matched rows from a 3,348-record high-risk subset). Every entry has a CDN screenshot (no TOR), normalized metadata, and a stable proof_id for reproducibility. We focus on (1) contact channels, (2) pricing visibility and currencies, and (3) volume over time.

Methodology

• Dataset: Illicit Goods — Q3 2025 (darknet marketplaces). Filtered to “precursor” mentions using a controlled keyword list (e.g., precursor(s), ephedrine, pseudoephedrine, safrol(e), BMK/PMK, P2P/MDP2P, red phosphorus, nitroethane, acetic anhydride, phenylacetic).
• Evidence model: Each row includes screenshot_url (HTTPS CDN) and proof_id (SHA-256 of normalized URL) to support auditability.
• Scope: OSINT only; no interactions or purchases; screenshots accessed safely via CDN.

Key Findings (Q3 2025)

1) Contact channels dominate XMPP.
XMPP ≈ 61.3% (1,572/2,565), Email ≈ 38.5% (988/2,565), onion handles ≈ 0.2% (5/2,565).
Implication: XMPP remains the default high-risk contact layer; email is the secondary channel.

2) Pricing visibility is partial but meaningful.
Non-zero price captured in ~36.9% of rows (947/2,565). This supports coarse trend analysis despite missingness.

3) USD is the reporting currency of record.
USD appears in ~88.2% of rows with a currency tag (2,262/2,565); missing currency tags ~11.8% (303/2,565).

4) Volume over time (Q3).
(Insert line chart from precursors_by_month.csv — counts by YYYY-MM.)
Note: Use counts as a proxy for offer momentum; seasonality and crawl cadence are discussed in limitations.

Charts & Tables

• Figure 1: Precursor mentions by month (precursors_by_month.csv) — line chart.
• Figure 2: Contact channels share (precursors_contact_types.csv) — bar chart.
• (Optional) Table: top contact handles or vendor_name signals (requires further parsing; out of scope for this snapshot).

Limitations

• Keyword-based matching may include broad “market category” pages alongside specific offers.
• Prices may be listed as ranges or in-text; missingness can bias medians.
• No MOQ normalization in this snapshot; a follow-up will extract MOQ phrases (“MOQ”, “minimum order”, “bulk”, “kg/L”) from text_snippet.

Ethics & Lawful Use

This analysis relies solely on publicly accessible OSINT and screenshot evidence. It does not endorse or facilitate illegal activity. Access via CDN removes the need for TOR in analytic environments.

Appendix: Reproducibility

Key fields: url, title, text_snippet, contacts, contact_type, price, currency, screenshot_url, proof_id, timestamp_utc.
Evidence: screenshots over HTTPS CDN; IDs via proof_id and ZIP checksums.

---

Dataset: https://reestrintelligence.gumroad.com/l/high-risk-goods-q3-2025?utm_source=hashnode&utm_medium=blog&utm_campaign=q3_2025
Affiliates: https://reestrintelligence.gumroad.com/affiliates

Contact

• Email: natalliavasilyeva777@gmail.com
• Website: reestr.blog
• X (Twitter): @reestr_ai
• LinkedIn: Natallia Vasilyeva
• Gumroad: reestrintelligence.gumroad.com
• Telegram: @reestr_global
• Affiliate program: Join here

Evidence-based intelligence for compliance & risk teams.