agent-loop-ext
Crash-resilient external agent loop with state persistence and CI/CD integration
Review an immutable dataset plan, impacts, capability degradation, and approval requirements before execution.
$ npx -y skills add jmagly/aiwg --skill dataset-plan-review --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/dataset-plan-reviewContext preview
The summary Claude sees to decide when to auto-load this skill.
Review an immutable dataset plan, impacts, capability degradation, and approval requirements before execution.
namespace: aiwg name: dataset-plan-review description: Review an immutable dataset plan, impacts, capability degradation, and approval requirements before execution. version: 1.0.0 platforms: [all] aliases: [dataset approval, ingest plan review, review indexing plan] triggers: [review the dataset plan, approve this ingestion, check write impacts, review indexing changes]
Use `aiwg dataset show <plan-ref> --json` and verify its digest, schemas, capability negotiation, artifact classes, estimated reads/writes, privacy, locality, network access, retention, authorization references, fallback, and rollback/reconciliation behavior. A preview is not execution authority.
Policy-sensitive writes require Dataset Policy Reviewer approval scoped to the exact plan digest. Retirement additionally requires complete enumeration and a reviewed bulk threshold. Emit a `dataset-workflow-handoff/v1` approval or rejection reference; never execute the plan here.
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