AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Gdelt Doc Search

skill-tiangong-ai-agent-skills-gdelt-doc-search · by tiangong-ai

Search GDELT DOC 2.0 through the Tiangong CLI for bounded article metadata, timeline aggregates, or tone-distribution bins using linted GDELT query syntax and optional split domain filters. Use for news-source discovery and trend reconnaissance; do not use for article bodies, raw GDELT event/GKG/mention rows, representative media measurement, fact verification, sentiment ground truth, or causal i…

— No reviews yet
0 installs
0 views
— view→install

Install

$ agentstack add skill-tiangong-ai-agent-skills-gdelt-doc-search

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • ✓ Prompt-injection patterns
  • ✓ Secret / credential exfiltration
  • ✓ Dangerous shell & filesystem operations
  • ✓ Untrusted network calls
  • ✓ Known-malicious package signatures

What it can access

  • ✓ Network access No
  • ✓ Filesystem access No
  • ✓ Shell / process execution No
  • ✓ Environment & secrets No
  • ✓ Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-tiangong-ai-agent-skills-gdelt-doc-search)

Reliability & compatibility

✓ Security review passed
0 installs to date
— no reviews yet
● 12d ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Gdelt Doc Search? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

GDELT DOC Search

Use the CLI-owned gdelt.doc-search capability. This Skill supplies intent routing and result-use boundaries only; the CLI owns source discovery, input/output schemas, HTTP behavior, limits, validation, and receipts.

Before running

  1. Read references/tiangong-data-requirement.json.
  2. Use the caller- or workspace-resolved stable CLI. The requirement declares

compatible capability and operation contract majors; it does not select a package build.

  1. Run data describe with that same CLI. Continue only when the capability

ID and required contract majors match, and copy the exact current capability/operation versions from that response into the run request.

  1. Immediately inspect manifest.availability and discovery.availability.

If either does not report available, stop before constructing a request and report its exact reasonCode, description, and resume criteria. Do not bypass suspension with a direct provider request. Route to GDELT Web NGrams only when the caller accepts its different literal phrase-search semantics.

tiangong-ai data describe gdelt.doc-search --json

Use the returned Discovery Metadata to confirm current source coverage, freshness, restrictions, provides, and doesNotProvide. Do not substitute facts remembered from an older Skill revision.

Prepare the request

Build a tiangong.data.run-request.v1 envelope. Replace the two version placeholders with the exact versions from the same data describe response. This example requests an article list for one bounded absolute UTC window:

{
  "schemaVersion": "tiangong.data.run-request.v1",
  "capabilityId": "gdelt.doc-search",
  "capabilityVersion": "",
  "operationId": "search",
  "operationVersion": "",
  "input": {
    "query": "(\"climate change\" OR pollution) sourcecountry:us",
    "mode": "artlist",
    "absoluteWindow": {
      "from": "2026-03-01T00:00:00Z",
      "to": "2026-03-07T23:59:59Z"
    },
    "maxRecords": 75,
    "sort": "hybridrel"
  }
}

Use the operation input schema returned by the same data describe response to select a supported mode and its mode-specific fields. Use at most one time-window form; omitting both uses the provider default window. Never silently widen a supplied window or pass arbitrary DOC parameters absent from the schema.

The CLI lints every effective query before network access. Wrap boolean OR groups in parentheses and use exactDomains (domainis:) or domains (domain:), never site: or inurl:. Repeated domains become separate bounded batches. Set continueOnQueryError only when a partial merged result is useful and retain batchQueries plus queryErrors.

Run

tiangong-ai data run gdelt.doc-search search \
  --input /absolute/path/to/request.json --json

The command emits a tiangong.data.run-result.v1 envelope. Preserve its contract, warnings, errors, and receipt with data when handing the result to another workflow.

Result boundaries

  • Persistent DOC throttling is unavailable retrieval, not zero coverage. When

literal topic discovery fits the task, consider the separately described gdelt-web-ngrams-search Skill as an explicit change of retrieval method. It does not support DOC operators or reproduce DOC aggregate timelines.

  • Treat article-list results as source metadata and links, not downloaded

article bodies, verified claims, or independent evidence units.

  • Treat timelines, tone, language, and source-country outputs as automated

aggregates whose meaning depends on the selected mode; do not compare unlike measures or infer causality from them.

  • In tonechart, articleCount is the count assigned to a tone bin, not the

tone score itself and not a public-sentiment estimate.

  • GDELT source coverage and automated extraction can be uneven. Do not treat

counts, tone, or rankings as representative measurements of public opinion, media prevalence, event truth, or sentiment ground truth.

  • Surface partial, truncation warnings, and empty results. Never reinterpret

them as complete absence of coverage.

  • Use a dedicated GDELT Events, GKG, or Mentions Skill when structured feed rows

are required: GKG supplies document annotations, Events supplies coded events, and Mentions links events to documents. This Skill must not invoke or combine them automatically, and their statistics are not interchangeable with DOC.

  • Cross-source comparison, full-text acquisition, persistence, polling, and

research evidence admission belong to the caller or Auto Research.

Reference

  • references/tiangong-data-requirement.json: stable capability requirement; it is not a package lock.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.