OnepointfiveHz/g8_100_tsk

jun271000g8100

Lossless, restorable payload archives for the selected task data in the jun_27_1000_g8_100 batch.

Contents

• —jun_27_1000_g8_100__pure_harbor_tasks.tar.zst: 92 rollout-ready Harbor tasks and their complete raw skill inputs. Pipeline intermediates, rollout/judge outputs, .setup_synthesis, nested .git metadata, and macOS Finder/AppleDouble files are excluded.
• —jun_27_1000_g8_100__rubrics.tar.zst: 176 validated final rubric JSON files (88 completion and 88 process), stored separately from the pure task archive.
• —selection_manifest.json: exact selection rules, task IDs, exclusions, rubric coverage, and migrated-status advisory.
• —archive_manifest.json: archive roots, sizes, checksums, member summaries, compression, and metadata normalization.
• —verification_report.json: canonical dry-run, rubric validation, source-stability, extraction, path, content, and metadata checks.
• —pure_tasks_source_snapshot.jsonl.zst and rubrics_source_snapshot.jsonl.zst: compressed source snapshots used for provenance and restoration verification.
• —SHA256SUMS: SHA-256 checksums for all uploaded data and metadata artifacts.

The main archive restores these roots:

raw_skills/jun_27_1000_g8_100/
synthesized_data/jun_27_1000_g8_100/
pipeline_runs/jun_27_1000_g8_100/

The pipeline_runs batch directory in the main archive is intentionally empty. The separate rubric archive restores validated rubric files beneath it.

Local macOS com.apple.FinderInfo, com.apple.quarantine, and com.apple.provenance extended attributes are transport metadata, not task payload, and are intentionally normalized out. File contents, paths, symlink targets, modes, uid/gid, and nanosecond mtimes are verified for the selected payload.

Verify and restore

From this repository directory:

sha256sum -c SHA256SUMS
mkdir -p /path/to/restore
zstd -dc jun_27_1000_g8_100__pure_harbor_tasks.tar.zst | tar -xf - -C /path/to/restore
zstd -dc jun_27_1000_g8_100__rubrics.tar.zst | tar -xf - -C /path/to/restore

The archives were verified by full extraction, exact path-set comparison, file size and SHA-256 comparison, normalized metadata comparison, validation of every final rubric, and a canonical rollout dry-run precheck for all 92 selected tasks. See verification_report.json for the recorded checks.