fact-check-event
Trace Digital SZTU event claims to source locators, compare support and counterevidence, check time and identity uncertainty, and produce a fact-check matrix.
Skill by Shuang-su
Chalk Labs
Need help deploying fact-check-event?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/shuang-su-fact-check-event
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/fact-check-event
What's inside
- Skill name
fact-check-event- Description
- Trace Digital SZTU event claims to source locators, compare support and counterevidence, check time and identity uncertainty, and produce a fact-check matrix. Use for 核查, 纠错, 争议, duplicate-event, or provenance questions; remain read-only unless the user asks to apply corrections.
- License
- Apache-2.0
- Instructions
- 17 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Fact-check an event 1. Work in the user-selected Digital SZTU clone. If an Event exists, read it, every cited Source, linked Event/Node, and generated backlinks. If the user only supplies a statement or provenance question, give a limited initial assessment from the available material; do not invent records or locators to satisfy this workflow. Use the pinned local CLI environment from `README.md` only when a CLI check is needed. 2. For each Claim report: text, kind, certainty, Event time, supporting Source locators, contradicting Source locators, independence groups, and unresolved questions. 3. Count only `supports` Citations from different non-empty independence groups as corroboration. Do not count copies, context, counterevidence, OCR, AI summaries, or vector search results as independent sources. For chat Sources, read [the chat workflow](../../docs/CHAT_IMPORT.md). Check the supplied export against its recorded SHA-256 when the original is available, then resolve each citation to the original message, time and surrounding context. Distinguish server IDs, database-local IDs and exporter-generated indexes; keep large IDs as strings. A missing original is an evidence gap, not a reason to invent a locator. Re-exports of one conversation belong to the same independence group. 4. Verify that Event-to-Node relationships use the right identity. Similar names or aliases do not prove they are the same person, organization, place, or institution. 5. Preserve conflicting evidence. “Not yet confirmed” is not the same as “disproved.” 6. When asked to apply a correction, edit the smallest Claim/Citation/Link scope, explain the evidence change, rebuild, and keep Git history intact. Do not assign governance roles or turn a fact-check report into an authority-based publication decision. Keep three findings separate: tool provenance, continuity of the supplied export's bytes, and the truth of a message's assertion. A matching tool/archive checksum or an upstream “Safe” name does not prove either chat authenticity or a Claim. Record actual inspection and untested assumptions explicitly; do not require access to an account or private database to complete a limited, read-only evidence report. For an onboarding suggestion, use [the shared exploration flow](../../docs/GETTING_STARTED.md#5-探索材料与历史线索) for scope and reporting, and assess only the available evidence. A Computer History observation should lead back to the original page, file or message; a retained summary is not independent corroboration. Distinguish event, message, publication, export, file-metadata and observation times rather than filling an unknown event date from the time someone opened a file.
From Shuang-su/digital-sztu, Apache-2.0.
More from Shuang-su/digital-sztu
- 0.564score
record-campus-event
Shuang-su
Create or update one source-linked Digital SZTU event from user-provided text, files, URLs, or original chat exports, then associate it with people…
SkillBFreshness: B★ 3Claude Code +9 - 0.561score
map-chat-to-events
Shuang-su
Help choose a pinned third-party chat export tool or use a user-selected export as a traceable Digital SZTU Source; map to chat JSONL and candidate Events…
SkillBFreshness: B★ 3Claude Code +9 - 0.559score
setup-digital-sztu
Shuang-su
Set up, resume, or check a local Digital SZTU working copy, then help find a first campus recording idea through authorized material and history exploration.
SkillBFreshness: B★ 3Claude Code +9 - 0.559score
compose-history
Shuang-su
Arrange existing source-linked Digital SZTU events as 编年体, 纪传体, 典制体, 纪事本末体, or another focused collection without duplicating event facts.
SkillBFreshness: B★ 3Claude Code +9 - 0.556score
link-event-directory
Shuang-su
Inspect, create, or repair Digital SZTU event relationships and Obsidian-style backlinks across event, person, organization, place, institution, topic…
SkillBFreshness: B★ 3Claude Code +9 - 0.556score
export-knowledge-base
Shuang-su
Build and validate deterministic, provider-neutral JSONL from Digital SZTU events, claims, nodes, sources, collections, and backlinks for full-text, graph…
SkillBFreshness: B★ 3Claude Code +9 - 0.553score
setup-sztu-connect
Shuang-su
Compatibility entry for initializing or checking Digital SZTU using the former SZTU Connect name.
SkillBFreshness: B★ 3Claude Code +9
Signals
- Stars
- 3
- Forks
- 0
- Open issues
- 5
- Last commit (this folder)
- Sep 5, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- Apache-2.0
- Repository
- github.com/Shuang-su/digital-sztu
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from Shuang-su
- 0.564score
record-campus-event
Shuang-su
Create or update one source-linked Digital SZTU event from user-provided text, files, URLs, or original chat exports, then associate it with people…
SkillBFreshness: B★ 3Claude Code +9 - 0.561score
map-chat-to-events
Shuang-su
Help choose a pinned third-party chat export tool or use a user-selected export as a traceable Digital SZTU Source; map to chat JSONL and candidate Events…
SkillBFreshness: B★ 3Claude Code +9 - 0.559score
setup-digital-sztu
Shuang-su
Set up, resume, or check a local Digital SZTU working copy, then help find a first campus recording idea through authorized material and history exploration.
SkillBFreshness: B★ 3Claude Code +9 - 0.559score
compose-history
Shuang-su
Arrange existing source-linked Digital SZTU events as 编年体, 纪传体, 典制体, 纪事本末体, or another focused collection without duplicating event facts.
SkillBFreshness: B★ 3Claude Code +9 - 0.556score
link-event-directory
Shuang-su
Inspect, create, or repair Digital SZTU event relationships and Obsidian-style backlinks across event, person, organization, place, institution, topic…
SkillBFreshness: B★ 3Claude Code +9 - 0.556score
export-knowledge-base
Shuang-su
Build and validate deterministic, provider-neutral JSONL from Digital SZTU events, claims, nodes, sources, collections, and backlinks for full-text, graph…
SkillBFreshness: B★ 3Claude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get fact-check-event running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
fact-check-event skill. Agent Mart. https://agentmart.dev/skills/shuang-su-fact-check-event (last updated 2026-10-24)