Batch Issues Preparation — classify, DAG, compose batches, enqueue
Objective: Turn an issue list/range into admitted, classified, batched scheduler queue entries: free batch compose over the whole set first, body-readiness screen + backfill to min_members, decision-grade classify only admitted members, risk-floor, deploy-pipeline and >8-file issues as review=full members (at most 2 per batch), no batch under 2 members, then anchor, audit, claim and enqueue with a per-member review level
Use it
Inspect metadata
ai-dossier info imboard-ai/git/batch-issues-preparation Download to the current directory
ai-dossier pull imboard-ai/git/batch-issues-preparation Verify, then run
ai-dossier run imboard-ai/git/batch-issues-preparationNeeds the CLI.
Trust and risk
| Declared risk level | medium (network_access) |
|---|---|
| Requires approval | no |
| Checksum | sha256:a1f16f9315a443fa6689bb393c7202dd63eadb3825df5760311e3533beb619e4 |
| Signature | ed25519, key imboard-ai by Yuval Dimnik <yuval.dimnik@gmail.com> |
| Signed at | 2026-09-24T16:01:50.314Z (covers frontmatter+body) |
| Public key | m97FPrnq/zKlQArLvJl3bTZCUMWWpp/d0UJ/OfUKZeE= |
These values are copied from the published dossier header. This site does not verify
them: run ai-dossier verify before trusting a dossier.
Details
| Categories | development |
|---|---|
| Tags | issue, workflow, batch-cycles, classification, scheduling, runstate, batch-compose |
| Authors | Yuval Dimnik |
| Source | raw .ds.md |