Enterprise data analysis
[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.From the repository description
npx -y skills add cxcscmu/SkillLearnBench --skill enterprise-data-analysisAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
SKILL.md
1.5 KB, 322 tokens by cl100k_base, as published. Nobody here has run it
name: enterprise-data-analysis description: Analyzing relationships between enterprise data entities (employees, products, reports, teams). Use this skill when a question requires connecting different types of information.
Enterprise Data Analysis
This skill provides a framework for understanding and navigating complex relationships in enterprise datasets.
Relational Mapping
- Product to Team: Look for a "team" or "team_members" field in the product JSON. If it's just names, you'll need to look up their IDs in the
employee.jsonfile. - Product to Competitor: Products usually have a
competitorsarray. This array may contain objects with details like names, strengths, weaknesses, and URLs. - Report to Author/Reviewer: Reports often have
authorsandreviewersfields. These are usually lists of names or IDs. - Insights to Employee: Insights about competitors or products may be linked to specific team members who provided them. Check for "provided_by", "member", or "author" fields within feedback or insight arrays.
Linking Data
- Identify the Source: Determine which product file contains the core information (e.g.,
ContentForce.json). - Extract Primary Keys: Get names or IDs of the relevant people/competitors from the product file.
- Cross-Reference: If only names are found, look up the
employee.jsonto find corresponding IDs. - Consolidate: Bring the findings together to form a complete answer (e.g., a list of IDs).
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.