Get card
Download
ingest-stage@1.0.0.yamlClone
npx -y darkprint clone ingest-stage@1.0.0There is no repository and no history behind a card. Download hands you the document as it stands, and Clone fetches the same document by name.
Run the first stage of the plan: pull every source the job names, reject what the plan does not ask for, and emit the staged working set together with the manifest of what was read.
used in 1 blueprint
Specification
111 words · handed to the agentExecute the first stage of the plan you are handed and nothing further. Pull every source that stage names, 250 records to a batch, and abandon the stage if any single batch takes longer than 180 seconds; pull nothing the stage does not name, however relevant it looks. Record for each named source what you read from it and when, or that it was unreachable and why. Emit on staged the working set you assembled, that manifest, and the stages still to run, so the next stage picks up where you stopped. You are done when every source the stage names appears in the manifest under one of those two outcomes.
Interfaces
1 in · 1 outInputs
1| Name | Data type | Required | Description |
|---|---|---|---|
| stages | plan | required | The full stage plan; this node executes the first stage and passes the rest along. |
Outputs
1| Name | Data type | Description |
|---|---|---|
| staged | json | The staged working set plus the source manifest the later stages are checked against. |
Dependencies
1The upstream nodes this card expects to receive from. Whenever a blueprint pins this card, each name is checked against a real edge in that graph. A name without a link is not a published card; it refers to a node inside some graph.
Card values
16 declaredWho the node is. The id is the key the DOT pins.
- id
- ingest-stage
- name
- Ingest Stage
- type
- agent
- phases
- Implementation
What it does, and the prose the agent is handed when the graph runs.
- action35 words
- Run the first stage of the plan: pull every source the job names, reject what the plan does not ask for, and emit the staged working set together with the manifest of what was read.
- spec111 words
- Execute the first stage of the plan you are handed and nothing further. Pull every source that stage names, 250 records to a batch, and abandon the stage if any single batch takes longer than 180 seconds; pull nothing the stage does not name, however relevant it looks. Record for each named source what you read from it and when, or that it was unreachable and why. Emit on
stagedthe working set you assembled, that manifest, and the stages still to run, so the next stage picks up where you stopped. You are done when every source the stage names appears in the manifest under one of those two outcomes.in full above - model
- claude-haiku-4-5
- agent
- Stage workers
- skill
- skills/ingest-stage.md
- tools
- none
- mcp
- none
- params
- timeout_s: 180, batch_size: 250
What arrives, what leaves, which nodes it expects to hear from, and what may not.
- inputs
- stages : plan
- outputs
- staged : json
- dependencies
- stage-planner
- cannot
- no type is refused
- will_not
- pull a source the stage does not name
The keys the static analysis reads. Nothing here instructs the agent.
- risk_markers
- none
- notes
- none
The card's own version, and who wrote it.
- version
- 1.0.0
- author
- autogen
- provenance
- not stated
Version history
1 version published- ingest-stage@1.0.0currentsha256:19a4cb823f62200a135230fb06582eafc14a7e49b239885af3002f68c89af5e5
pinned byCheckpoint & Resume Runner
autogen/checkpoint-resume-runner
A digest is a fingerprint (SHA-256) of the card's content, computed without the author and provenance fields. The same card from two people gets the same digest; any edit gets a new one.
First published version, so there is nothing to compare yet. Versions are never edited in place: the next change arrives as a new version, and the differences between the two documents are listed here.
Community notes (0)
No notes yet.
Nobody has posted about this node card yet.
Sign in to post a note.