Menü
Sync Delivery Graph nodes to external task trackers such as Linear, Azure DevOps, or GitHub Issues.
Mit Codex oder Claude installieren Kopieren Sie diesen Prompt, fügen Sie ihn in Codex, Claude oder einen anderen Assistant ein und lassen Sie die Skill-Seite prüfen und installieren.
Basierend auf der SOC-Berufsklassifikation
Capture reusable learning from completed Delivery Graph nodes, tracks, or demands.
Drive a raw demand all the way to done from a single command - design, plan, and evidence-gated execution - stopping only at two judgment gates and on genuine failure or ambiguity.
Design a structured demand from a raw request, including its gap register, requirements, and acceptance criteria for Delivery Graph Engineering.
Drive the ready queue end to end - implement each ready node, gate it on evidence, and stop on the first failure.
Convert DGE requirements into tracks, dependency nodes, and validation contracts.
Independently verify a Delivery Graph node against its validation contract and block completion until evidence and verifier independence exist.
| name | dge-sync |
| description | Sync Delivery Graph nodes to external task trackers such as Linear, Azure DevOps, or GitHub Issues. |
| argument-hint | [linear|ado|github|all] [optional graph path] |
Use this skill when the graph should create or update tracker records.
Project the canonical delivery graph into external systems. Trackers are projections; the graph remains the source of truth.
delivery-graph/sync/<target>.json after successful sync.| DGE object | Linear | Azure DevOps |
|---|---|---|
| Demand | Project / Initiative | Feature / Epic |
| Requirement | Milestone, label, or parent issue | User Story / PBI |
| Track | Project view, label, or cycle | Area grouping |
| Work node | Issue | Task |
| Dependency | Blocks relation | Related/predecessor link |
| Validation | Checklist/comment | Acceptance checklist |
delivery-graph/graph.json.delivery-graph/sync/.Follow the shared output convention (see skills/README.md): lead the final
reply with a bold one-line synthesis (e.g. "Synced: 3 created, 2 updated, 1
conflict"), then the created/updated/skipped/conflicted record list, then a
## Next block: resolve reported conflicts, or /dge-work-node for
newly-synced items.
Use local dry-run adapters first:
npm run dge -- sync linear --team-id "<linear-team-id>"
npm run dge -- sync ado --org "<ado-org>" --project "<ado-project>" --area "<area-path>" --iteration "<iteration-path>"
These write delivery-graph/sync/linear.json or delivery-graph/sync/ado.json with planned tracker payloads. Treat those files as the reviewable projection before enabling any API-backed sync.
If credentials or tracker tools are unavailable, write a dry-run plan instead of failing silently:
delivery-graph/sync/<target>.json