Safe Mass Index Core

Safe Mass Index Core

Use this skill to index large repositories safely without rg subprocess churn.

Workflow

1. Set bounded policies (time/files/read-bytes, exclusions).
2. Build incremental or full metadata index.
3. Query by path/ext/lang/size/freshness/scope.
4. Use run artifacts to verify deterministic bounded execution.

Policy

1. Do not use rg/rg.exe for mass indexing.
2. Use only local filesystem metadata and bounded lightweight probing.
3. Keep runs bounded with file/time/read-byte budgets.

Build Command

python3 "$CODEX_HOME/skills/safe-mass-index-core/scripts/index_build.py" \
  --repo-root <path> \
  --index-dir .codex-index \
  --mode incremental \
  --max-files-per-run 12000 \
  --max-seconds 25 \
  --max-read-bytes 67108864 \
  --exclude-dir .git \
  --exclude-dir node_modules \
  --exclude-dir .venv \
  --exclude-dir venv \
  --exclude-dir __pycache__ \
  --exclude-dir dist \
  --exclude-dir build \
  --exclude-dir target \
  --exclude-dir .cache \
  --exclude-dir .codex-index \
  --json-out .codex-index/run.json

For very large repos, enable sharded storage:

python3 "$CODEX_HOME/skills/safe-mass-index-core/scripts/index_build.py" \
  --repo-root <path> \
  --index-dir .codex-index \
  --mode incremental \
  --max-files-per-run 12000 \
  --max-seconds 25 \
  --max-read-bytes 67108864 \
  --sharded

Query Command

python3 "$CODEX_HOME/skills/safe-mass-index-core/scripts/index_query.py" \
  --index-dir .codex-index \
  --path-contains <text> \
  --ext <extension> \
  --lang <language_tag> \
  --min-size <bytes> \
  --max-size <bytes> \
  --changed-since-epoch <seconds> \
  --scope <top_level_dir_or_all> \
  --limit 200 \
  --format table

Outputs

• .codex-index/manifest.json
• .codex-index/state.json
• .codex-index/files.jsonl (non-sharded)
• .codex-index/shards/*.jsonl (sharded)
• .codex-index/run.json

Scope Boundary

Use this skill only for the safe-mass-index-core lane and workflow defined in this file and its references.

Do not use this skill for unrelated lanes; route those through $skill-hub and the most specific matching skill.

Reference

• references/index-schema.md

Loopback

If this lane is unresolved, blocked, or ambiguous:

1. Capture current evidence and failure context.
2. Route back through $skill-hub for chain recalculation.
3. Resume only after the updated chain returns a deterministic next step.