メニュー
Close implementation loops with target-bound independent evidence and a portable proof receipt. Use when an agent changes a real system and must earn a completion claim that survives the session.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
| name | finish-the-loop |
| description | Close implementation loops with target-bound independent evidence and a portable proof receipt. Use when an agent changes a real system and must earn a completion claim that survives the session. |
Treat completion as a claim that needs evidence from the intended target.
WORKFLOW.json and create .skdd/runs/<run-id>/state.json before changing the system.receipt.json using the proof-receipt schema. Bind the skill bytes and each acceptance contract with SHA-256 digests.skdd proof verify receipt.json --skill SKILL.md.verified only when the verifier passes. Otherwise report the reason codes and leave the loop refused, blocked, or open.Never translate “the build passed” into “the user-visible outcome works.” Evidence must match the acceptance target exactly.
First fixture skill for skdd add tests. Use when testing drop installation.
Second fixture skill for skdd add tests. Use when testing multi-skill drops.
Safely backfill historical data for a pipeline stage — scope the date range, run a dry-run, execute in chunks to avoid overwhelming the database, and verify row counts after each chunk. Use when historical data is missing, when a pipeline stage is added retroactively, or when asked to "backfill the {stage} table from {start_date}".
Add or extend data validation checks in the pipeline when data quality issues are detected — null-percentage thresholds, value ranges, referential integrity, and row-count sanity checks. Use when a pipeline breaks on bad data, when onboarding a new dataset, or when asked to "add a quality check for {column}".
Onboard a new raw dataset into the data platform — sniff the schema, generate a profiling notebook, create the ingestion job, and add an entry to the data dictionary. Use when adding a new data source, when a partner delivers a new CSV/Parquet drop, or when asked to "onboard the {name} dataset".
Record a model training run — log parameters, metrics, and artifact paths, then append a structured row to the experiments ledger. Use when a training run completes, when asked to "log this experiment", or when comparing runs to decide which model to promote.