ClawHub

Screen recommendation loop

v0.0.1

Build and run a low-friction movie/anime recommendation + follow-up loop. Use when a user wants long-term taste profiling from watched/unfinished/dropped fee...

⭐ 0· 180·0 current·0 all-time
by@gloryxia
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name and description match the SKILL.md: the document describes a low-friction recommendation loop, candidate pools, follow-up timing, a small record schema, and update heuristics. It does not request unrelated credentials, binaries, or install actions.
ℹ
Instruction Scope
Instructions remain focused on recommendation/polling and local state (JSON/SQLite). They assume the agent/platform can send messages and schedule follow-ups; the SKILL.md does not prescribe network access or APIs, but it references external candidate pools (Douban/Bangumi) without specifying how to fetch them. If an implementation adds scraping/API calls, that expands scope and may require credentials or further privacy review.
✓
Install Mechanism
There is no install spec or code; this instruction-only skill writes nothing to disk by itself. That minimizes install-time risk.
✓
Credentials
The skill declares no environment variables, credentials, or config paths. Its data storage suggestion (JSON or SQLite) is proportionate to the stated purpose. Be aware that a concrete implementation that pulls from external services may later require API keys.
✓
Persistence & Privilege
always is false and autonomous invocation is not disabled (the platform default). The skill does not request permanent platform-level presence or access to other skills' configs.
Assessment
This instruction-only skill is coherent and lightweight, but before installing consider: where will you store the JSON/SQLite state (ensure appropriate access controls/backups)? If you or an implementer add fetching from Douban/Bangumi or other external sources, review what network access and API credentials are needed and whether scraping is allowed. Also confirm the agent/platform's scheduling and message-sending behavior so follow-ups behave as you expect and no private identifiers (chat IDs, account tokens) are recorded by the skill.

Like a lobster shell, security has layers — review code before you run it.

latestvk9720nevw59rm79epfjx99m5q182p2re

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Comments