agent-loop-ext
Crash-resilient external agent loop with state persistence and CI/CD integration
Delegate approved one-shot or incremental dataset execution to the shared orchestration service with resumable receipts.
$ npx -y skills add jmagly/aiwg --skill dataset-ingest --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/dataset-ingestContext preview
The summary Claude sees to decide when to auto-load this skill.
Delegate approved one-shot or incremental dataset execution to the shared orchestration service with resumable receipts.
namespace: aiwg name: dataset-ingest description: Delegate approved one-shot or incremental dataset execution to the shared orchestration service with resumable receipts. version: 1.0.0 platforms: [all] aliases: [run dataset plan, materialize dataset, resume ingest, synchronize dataset] triggers: [ingest the dataset, run the approved plan, resume dataset sync, update this index, continue ingestion]
Require the approved immutable plan reference and matching approval reference. Delegate only through `aiwg dataset ingest <plan-id> --digest <plan-digest> --idempotency-key <key> --json`, adding only reviewed `--approve` and reconciliation bindings. For an incremental continuation, use `retry <run-id>` when the orchestration service reports the run retryable; never decode or advance a cursor here.
Surface run ID, status, cancellation affordance, prior and new committed checkpoint references, rejections, degraded behavior, and receipt reference. Cancellation must preserve the last committed checkpoint. Hand the run/receipt references to `dataset-verify` and, when requested, `dataset-trace`.
Reusable project context and specialist workflows for the AI tools you already use. Plan software, coordinate specialist reviews, prepare campaigns, investigate incidents, organize research, curate media, and maintain operational knowledge.
Repo: jmagly/aiwg
Crash-resilient external agent loop with state persistence and CI/CD integration
Detect requests for iterative autonomous agent loops and route to the appropriate loop executor
Automatically execute tests when code-generating agents modify source files, enforcing the execute-before-return pattern
Enable agent loops to learn from similar past tasks and share patterns across loops
Query and manage the executable feedback debug memory
Execute tests on generated code and iterate until passing