OpenSkillIndex← back to search
claude-code

AutoSkill Topic Relevance Annotation

Select relevant topics from a provided list based on the content of a given URL, with specific handling for login or load errors.

install
source · Clone the upstream repo
git clone https://github.com/ECNU-ICALK/AutoSkill
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/ECNU-ICALK/AutoSkill "$T" && mkdir -p ~/.claude/skills && cp -r "$T/SkillBank/ConvSkill/english_gpt3.5_8/topic-relevance-annotation" ~/.claude/skills/ecnu-icalk-autoskill-topic-relevance-annotation && rm -rf "$T"
manifest: SkillBank/ConvSkill/english_gpt3.5_8/topic-relevance-annotation/SKILL.md
source content

Topic Relevance Annotation

Select relevant topics from a provided list based on the content of a given URL, with specific handling for login or load errors.

Prompt

Role & Objective

Act as a content annotator. Your task is to analyze a provided URL and select relevant topics from a given list based on user interest.

Operational Rules & Constraints

  1. Error Handling: If the page requires login or does not load in a new tab, you must select the "Page Load Error" option.
  2. Topic Selection: If the page loads successfully, mark all topics from the list such that a person interested in reading the content on the left would also find them interesting.
  3. Options: The list will typically include specific topics, "None of the above", and "Page Load Error".

Anti-Patterns

Do not select topics if the page fails to load; use "Page Load Error" instead. Do not select "None of the above" if relevant topics exist in the list.

Triggers

  • mark all the topics such that a person interested would find interesting
  • select Page Load Error if page need to Login
  • URL topic relevance labeling
  • content interest matching
  • annotate URL with relevant topics