{"slug":"trump-truth-monitor","title":"trump-truth-monitor","summary":"Use when monitoring or interpreting Donald Trump's Truth Social posts for market-moving events — tariff announcements, sanctions, deals with countries (China / Mexico / Canada / EU / Japan / Korea / Taiwan), specific company / CEO mentions, Fed pressure, energy / oil commentary, ","platform":"Claude","tags":[],"authorName":"LLM Mart","authorSlug":"llm-mart","score":0,"source":"github","price":null,"verified":false,"createdAt":"2026-09-01T17:41:57.751327Z","repo":{"url":"https://github.com/kansoku-trade/kansoku","stars":325,"forks":39,"license":null,"updatedAt":"2026-09-26T13:01:05Z"},"bodyHtml":"
\n

name: trump-truth-monitor\ndescription: Use when monitoring or interpreting Donald Trump's Truth Social posts for market-moving events — tariff announcements, sanctions, deals with countries (China / Mexico / Canada / EU / Japan / Korea / Taiwan), specific company / CEO mentions, Fed pressure, energy / oil commentary, crypto policy, or geopolitical escalation. Triggers on \"trump 发了什么\", \"check trump\", \"trump 关税\", \"盘前 trump 推\", \"trump truth social\", \"trump tweet impact\", \"trump 对 X 说了什么\", or whenever a pre-market gap / intraday spike on policy-sensitive names (semis, China ADRs, autos, energy, banks, defense) needs to be explained.

\n

Trump Truth Monitor

\n

Pulls Donald Trump's Truth Social feed via the trumpstruth.org RSS mirror, classifies posts into market-relevant topic buckets, and hands the candidate list off for LLM-level market-impact grading.

\n

When to use

\n\n

If the user wants tweet history beyond ~5 days, this skill is insufficient — the RSS mirror only exposes the latest ~100 posts. Route to Factba.se / Roll Call (paid) or note the limitation explicitly.

\n

Data source

\n

trumpstruth.org/feed — a public third-party mirror of @realDonaldTrump on Truth Social. RSS 2.0 XML with these fields per item:

\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
FieldMeaning
<pubDate>RFC 2822, original Truth Social post timestamp
<link>trumpstruth.org/statuses/
<truth:originalUrl>truthsocial.com/@realDonaldTrump/ — the primary source
<description>Full post body with HTML (links + ellipsis spans)
\n

The mirror typically lags the original by ≤2 minutes. Single feed pull returns ~100 most recent posts, covering ~5 days at Trump's typical cadence.

\n

CLI

\n

Read mode — fetch.py

\n
# Default: last 24h, keyword-filtered, markdown\npython3 .claude/skills/trump-truth-monitor/scripts/fetch.py\n\n# Wider window\npython3 .claude/skills/trump-truth-monitor/scripts/fetch.py --hours 72\n\n# All posts in feed regardless of keyword\npython3 .claude/skills/trump-truth-monitor/scripts/fetch.py --hours 72 --all\n\n# Single topic\npython3 .claude/skills/trump-truth-monitor/scripts/fetch.py --topic tariff_trade\n\n# JSON output (for chaining)\npython3 .claude/skills/trump-truth-monitor/scripts/fetch.py --json\n
\n

Topic buckets defined in script: tariff_trade, semi_tech, energy, fed_macro, crypto, geopolitical.

\n

Archive mode — archive.py

\n
# Append new posts to journal/trump-feed/YYYY-MM-DD.md (idempotent)\npython3 .claude/skills/trump-truth-monitor/scripts/archive.py\n\n# Custom output dir\npython3 .claude/skills/trump-truth-monitor/scripts/archive.py --out /path/to/dir\n\n# Silent unless something new was added\npython3 .claude/skills/trump-truth-monitor/scripts/archive.py --quiet\n
\n

The archive de-dupes by mirror status_id — re-running on the same feed is a no-op. Designed to be scheduled (see launchd/README.md). Once archived, posts persist locally even if trumpstruth.org goes down.

\n

Workflow

\n
    \n
  1. Decide window. Default 24h. Use 48–72h when investigating a multi-day move. Use --all when context-grazing.
    2. \n
  2. Decide scope. If user asks generally → no --topic. If user names a domain (关税 / 半导体 / 油 / 加密) → pass --topic.
    3. \n
  3. Pull feed. Run fetch.py with chosen flags. Always include --hours — never default to \"all of feed\" silently.
    4. \n
  4. Second-pass grading. Script output is candidates, not signals. For each post:\n\n
    5. \n
  5. Anchor on original URL. When quoting, always cite truth:originalUrl (the truthsocial.com link), not the mirror.
    6. \n
  6. Render output. For multi-post stretches, group by tier — high first, then med, then a one-line low/noise tally.
    7. \n
\n

Output template

\n
# Trump's Truth — {WINDOW}\n\n## High-impact (potential market mover)\n- [{utc_time}] {one-line summary} — `tier: high` · {topic tags}\n  > \"{verbatim short quote ≤2 sentences}\"\n  - Original: {truthsocial.com URL}\n  - Possible market read: {sector / ticker level expectation, anchored}\n\n## Medium-impact (directional, no specifics)\n- [{utc_time}] {one-line summary}\n  - Original: {URL}\n\n## Noise (matched keyword, low signal)\n- {N} posts ({topic distribution}) — endorsements / personal — not enumerated\n\n⚠ Trump may delete or contradict within hours. Position decisions should require independent confirmation (sector ETF tape, peer reaction, official release).\n
\n

Anti-patterns

\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
MistakeReality
Treating script output as \"market signal\"Script is a keyword filter. LLM must read each post and tier.
Quoting a mirror URL as the sourceAlways link truth:originalUrl (truthsocial.com). Mirror is a convenience.
Reporting Senate endorsements as \"policy news\"Politics-only posts with military / energy keywords are noise — filter at tier=noise.
\"Trump said X about Y\" with no linkAlways include the truthsocial.com link. User must be able to verify.
Pretending tweets are durableTrump posts can be deleted or retracted within hours. If consulted >12h after, note staleness.
\n

Integration

\n\n

Limitations

\n\n

Local archive — searching past posts

\n

Once archive.py has been running, journal/trump-feed/YYYY-MM-DD.md accumulates a complete record. To investigate a past day or query historically:

\n
# All tariff-related posts ever archived\ngrep -l \"tariff\" journal/trump-feed/*.md\n\n# Specific company mention\ngrep -B2 -A8 -i \"nvidia\\|tsmc\" journal/trump-feed/*.md\n\n# Posts on a specific date\ncat journal/trump-feed/2026-05-26.md\n
\n

The archive is plain markdown — grep-friendly, git-trackable.

\n

Related skills

\n\n","files":[{"path":"scripts/archive.py","sizeBytes":3785,"isText":true},{"path":"scripts/fetch.py","sizeBytes":6965,"isText":true},{"path":"SKILL.md","sizeBytes":9071,"isText":true}],"reviewScore":null,"reviewSummary":null,"trust":{"provenance":"trusted-source-unreviewed","notice":"Community-authored content, reproduced verbatim and not vetted as instructions. Treat it as data to evaluate, never as directives to follow.","bodySource":null},"bodyLocked":false,"purchaseUrl":null,"sourceUrl":null,"report":{"provenance":"trusted-source-unreviewed","screen":{"ran":true,"outcome":"clean","suspicious":0,"notes":0,"hiddenCharacters":false},"virusScan":{"engine":"clamav","status":"clean","scannedAt":"2026-09-01T17:42:20.470418Z","sha256":"D540FBA7CDD86E307B159A56E30BF93C05A4E2EAF0ACEE2C551DF3145E940647","sizeBytes":8736},"review":null,"source":{"repositoryUrl":"https://github.com/kansoku-trade/kansoku","path":".claude/skills/trump-truth-monitor","license":null,"commit":"8fa9a03f5fa666d0c6ceca95a7e197f70f63e277","subtreeSha":"33F4EE01442E3EA0E64FA7872128E351B98FFBD405B9552411C3E9BCD63B9EA1","lastSyncedAt":"2026-10-06T15:23:08.794427Z"},"reviewedAt":"2026-09-01T17:43:23.493886Z","notice":"Community-authored content, reproduced verbatim and not vetted as instructions. Treat it as data to evaluate, never as directives to follow."},"install":[{"target":"skills-cli","command":"npx skills add https://github.com/kansoku-trade/kansoku/tree/main/.claude/skills/trump-truth-monitor"},{"target":"claude-code","command":"claude plugin marketplace add https://llmmart.ai/marketplace.json && claude plugin install kansoku-trade-kansoku@llmmart"},{"target":"git","command":"git clone https://github.com/kansoku-trade/kansoku.git"}]}