agent-loop-ext
Crash-resilient external agent loop with state persistence and CI/CD integration
Recommend an independently composable dataset capability profile with plain-language rationale and safe defaults.
$ npx -y skills add jmagly/aiwg --skill dataset-capability-recommend --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/dataset-capability-recommendContext preview
The summary Claude sees to decide when to auto-load this skill.
Recommend an independently composable dataset capability profile with plain-language rationale and safe defaults.
namespace: aiwg name: dataset-capability-recommend description: Recommend an independently composable dataset capability profile with plain-language rationale and safe defaults. version: 1.0.0 platforms: [all] aliases: [indexing recommendation, dataset profile, retrieval plan] triggers: [how should I index this, make it searchable with provenance, add traceability, choose vector or graph, recommend dataset capabilities]
Consume intake and assessment references. Recommend only capabilities supported by current negotiation evidence. Compose full-text/vector/hybrid/rerank indexing, traceability, provenance, graph, and export independently; never force a preset.
For each selection state rationale, assumptions, materialized artifact and its class, privacy/network/write implications, required versus optional status, and explicit fallback. Required absence fails closed. Write those decisions to the governed plan input, then use `aiwg dataset plan --file <plan-input.json> --json` to create the candidate plan and hand its immutable reference to `dataset-plan-review`.
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