influx

High-signal X/Twitter influencer list — functional demo built on the open-source multi-agent framework CCCC. Ready-to-use whitelist of influential individual accounts (non-brand/non-official), curated for downstream ingestion. This open-source bundle is a download-and-use minimal set; the full production flow requires CCCC + RUBE MCP (Twitter tools) for data fetching.

Languages: English (this file) · 中文

What this is

• A rigorously filtered list of active, individual X.com authors (no brands/officials), aiming for 5k–10k high-value entries.
• Current release (404 records):
  • JSONL: data/release/influx-latest.jsonl
  • Gzipped: data/release/influx-latest.jsonl.gz
  • Manifest: data/release/manifest.json
• Delivered as data only; you don’t need to run the pipeline to use it.
• Components included here (minimal open-source set):
  • Data: data/release/ (latest JSONL + manifest)
  • Guard: scripts/pipeline_guard.sh
  • Schema: schema/bigv.schema.json
  • Rules: lists/rules/brand_heuristics.yml, lists/rules/risk_terms.yml
  • Sample prefetched JSONL: data/prefetched.sample.jsonl (for local filter demo)

Why it’s useful

• Content ingestion/ranking: high signal-to-noise whitelist reduces crawl and processing cost.
• Research/monitoring: supports trend tracking, community analysis, influence-network studies.
• Product bootstrapping: import a quality author set for recommendations, alerting, sentiment/market intel.

What’s in the list (selection criteria)

• Individuals only; brand/official/organization accounts excluded via heuristics.
• Thresholds: (Verified AND followers ≥30k) OR followers ≥50k.
• Recency: keeps recent activity fields (metrics_30d*); stale accounts are filtered upstream.
• Evidence: every record carries sources.evidence + fetched_at for auditability.
• Hard guards: handle and id globally unique; placeholder/“000” followers, mock/test prefixes, non-numeric IDs rejected; strict schema validation enforced.

Quick use

cp data/release/influx-latest.jsonl .
# or compressed
cp data/release/influx-latest.jsonl.gz . && gunzip influx-latest.jsonl.gz

import json

with open("influx-latest.jsonl") as f:
    authors = [json.loads(line) for line in f]

# Example: English AI authors, score >= 60
ai_authors = [
    a for a in authors
    if "ai_core" in a.get("topic_tags", [])
    and a.get("lang_primary") == "en"
    and a.get("score", 0) >= 60
]
print(len(ai_authors))

Data model (summary)

• Full schema: schema/bigv.schema.json
• Key fields: id (author_id), handle, name, verified, followers_count, lang_primary, topic_tags, metrics_30d*, meta.sources (with evidence/fetched_at), provenance_hash.

How it’s produced (context; requires external deps)

• Full flow requires: CCCC + RUBE MCP (Twitter tools) to fetch users → prefetched JSONL → run influx-harvest x-lists|bulk --prefetched-users <file> → enforce scripts/pipeline_guard.sh (dedup handle/id, evidence required, placeholder/“000” rejection, strict schema) → publish to data/release/.
• This repo ships the minimal set (data + guard + schema + rules + sample prefetched); MCP fetching is not included here.

License

• Apache-2.0 (covers code and released data).

Credits

• Built on the CCCC multi-agent framework; thanks to all contributors for curation, filtering, and QA.