score-rfi

npx skills add majiayu000/claude-skill-registry --skill score-rfi --agent claude-code

Same command for any agent — swap --agent for codex, cursor, copilot.

# /score-rfi Score vendor RFI responses stored in SQLite databases using a standardised 0-3 rubric. ## Usage ``` /score-rfi <vendor-name> /score-rfi pwc /score-rfi accenture /score-rfi ibm ``` ## Prerequisites Before scoring, ensure the vendor's CSV has been converted to SQLite: ```bash # Convert CSV to SQLite (see /csv-to-sql skill) node scripts/csv-to-sqlite.js "Inbox/<vendor>-rfi-responses.csv" --start-row <n> --fts --verbose ``` ## Scoring Rubric (0-3 Scale) | Score | Rating | Qualitative Assessment | | ----- | ------ | ---------------------------------------------------------------------------------------------------------------- | | **3** | High | Strong proven experience; clear, detailed answer demonstrating deep understanding; high confidence in capability | | **2** | Medium | Some capability demonstrated; potential but unproven or limited evidence; moderate risk | | **1** | Low | Insufficient evidence; generic or superficial response; does not address specifics; high risk | | **0** | Zero | Not demonstrated at all; no evidence or response; very high risk | ### Scoring Criteria When evaluating responses, consider: 1. **Domain Specificity** - Does the response demonstrate un

Convert CSV to SQLite (see /csv-to-sql skill)
node scripts/csv-to-sqlite.js "Inbox/<vendor>-rfi-responses.csv" --start-row <n> --fts --verbose
sqlite3 .data/<vendor>-rfi-scoring.db ".tables"
sqlite3 .data/<vendor>-rfi-scoring.db ".schema"
sqlite3 .data/<vendor>-rfi-scoring.db -markdown -header \
sqlite3 .data/<vendor>-rfi-scoring.db \
FROM <table> GROUP BY score ORDER BY score DESC;"
FROM <table>
WHERE <scorer_col> LIKE '1 -%' OR <scorer_col> LIKE '0 -%'
ORDER BY CAST(id AS INTEGER);"