Hotspot Research Suite

AI research agent for trend discovery, public-source analysis, and Lark-ready Chinese briefs.

Hotspot Research Suite helps researchers and deep writers find timely, evidence-backed, relatively low-competition topics before everyone else writes the same piece.

pip install --upgrade hotspot-research-cli
hotspot-research setup
hotspot-research brief "中文大模型安全评测的新兴低竞争切口" --field "中文大模型安全"

Use it when you need topic intelligence, must-read source lists, trend metrics, and a structured Markdown brief that can be sent to Feishu/Lark.

Why Star This Repo

• Turns recent public signals into concrete research and writing directions.
• Combines a reusable agent skill with a PyPI CLI.
• Uses Pydantic models and SQLite cache so outputs are structured and repeatable.
• Built for Chinese long-form research workflows and Lark distribution.

Related Repos

• hotspot-research-skill: standalone installable skill for long-form research reports.
• pain-miner: community pain-point mining for micro-product ideas.
• awesome-ai-agent-research-tools: the curated entrypoint for all related tools.

The current default product is hotspot-research-cli: an interactive topic intelligence assistant for researchers and deep writers. It helps users discover timely, evidence-backed, relatively low-competition writing and research topics, then saves a structured Chinese 《选题情报简报》 as Markdown.

Components

• skills/last30days-safe: safe public-source collection skill from the Hermes Agent skill library.
• packages/hotspot-cli: PyPI package hotspot-research-cli.

Install

pip install --upgrade hotspot-research-cli

Start the topic assistant:

hotspot-research setup
hotspot-research run --output-dir ./briefs

Validate an existing idea:

hotspot-research brief "中文大模型安全评测的新兴低竞争切口" --field "中文大模型安全"

Configure structured LLM analysis:

hotspot-research config model list
hotspot-research config model setup --provider deepseek --model deepseek/deepseek-chat
hotspot-research config lark auth --init --recommend --chat-id oc_xxxxxxxxx

What The CLI Produces

The default run flow asks for a field, gathers recent public signals through last30days-safe, proposes 5-8 concrete emerging directions, supports numeric selection or natural-language follow-up, then generates a Markdown brief with:

• timeliness and data signals
• current research coverage
• high-potential research gaps
• concrete writing/research questions
• title suggestions
• recent must-read materials
• risk notes
• trend metrics for 7-day, 30-day, and 30-60-day comparison windows

See packages/hotspot-cli/README.md for full CLI usage, architecture, Pydantic models, cache behavior, LLM configuration, and extension guidance.

Source Attribution

See NOTICE.md for source attribution and reference links.

• last30days-safe is copied from the Hermes Agent skill library and carries upstream MIT metadata.
• hotspot-cli was created locally for this suite.

Development

cd packages/hotspot-cli
python3 -m venv .venv
. .venv/bin/activate
python -m pip install -e .
PYTHONPATH=src python -m unittest discover -s tests -v

Publishing

This repo includes scripts/push-github.zsh, which reads the GitHub token from:

~/.config/hotspot-research-suite/github_token

The token file is local-only and must never be committed.

PyPI publishing uses GitHub Actions Trusted Publishing in .github/workflows/publish.yml. Push a path-scoped tag to publish:

git tag hotspot-research-cli/v0.2.0
scripts/push-github.zsh origin hotspot-research-cli/v0.2.0