# Pinned corpus for the Tier 1 retrieval-quality gate (scripts/eval_retrieval.py).
#
# The golden set (golden.jsonl) grades real catalog items by NAME, so the gate
# needs the repos that actually contain those items tapped before it can score.
# This is the minimal set of repos that covers all 77 golden targets across the
# skill / rule / workflow kinds. Over it, BM25 recall@10 == 1.000, so the 0.85
# floor gate has a wide margin and tolerates upstream drift.
#
# NOTE: `boost tap --defaults` is NOT sufficient — it taps only skill repos and
# scores 0.000 on every rule and workflow query (overall recall ~0.33). That is
# why this explicit list exists.
#
# If you edit this list, regenerate the baseline:
#   BOOST_HOME=/tmp/evalhome bash scripts/ensure_eval_corpus.sh
#   BOOST_HOME=/tmp/evalhome python3 scripts/eval_retrieval.py --build --save-baseline
#
# Future hardening: pin each repo to a commit SHA for full reproducibility.
anthropics/skills
0xfurai/claude-code-subagents
NeoLabHQ/context-engineering-kit
Aaronontheweb/dotnet-cursor-rules
ChrisWiles/claude-code-showcase
LessUp/awesome-cursorrules-zh
