Skip to content
DarkPrint
Aautogentransform-stage1.0.0

Transform Stage

Get card

Clone

npx -y darkprint clone transform-stage@1.0.0

There 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.

↓ 1 downloads

Apply the plan's transformation to the staged working set, or to the snapshot the resume controller hands back, and emit the transformed set with the diff against what it started from.

used in 1 blueprint

Specification

128 words · handed to the agent

You are handed a working set on staged on a first run, or a re-issued attempt on restore when a previous run of this stage failed and is being replayed; treat the two identically and prefer restore when both arrive. Apply the transformation the stage plan specifies to every record in that set, and abandon the stage if the work runs longer than 300 seconds rather than emitting a partial set. Emit on transformed the transformed set together with the diff against the input you started from, so the result can be replayed later without re-reading the sources. Be idempotent over your own output: run twice on the same input you must produce the same transformed set and the same diff, because a replayed attempt is exactly that.

Interfaces

2 in · 1 out

Inputs

2
Inputs declared by this node card
NameData typeRequiredDescription
stagedjson requiredThe working set the ingest stage produced, on a first run.
restorejson optionalThe attempt the resume controller re-issued from the last valid checkpoint.

Outputs

1
Outputs declared by this node card
NameData typeDescription
transformedjsonThe transformed working set, with the diff against its input so the checkpoint is replayable.

Dependencies

2

The 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

17 declared

Who the node is. The id is the key the DOT pins.

id
transform-stage
name
Transform Stage
type
agent
phases
Implementation
Behaviourcard spec →

What it does, and the prose the agent is handed when the graph runs.

action31 words
Apply the plan's transformation to the staged working set, or to the snapshot the resume controller hands back, and emit the transformed set with the diff against what it started from.
spec128 words
You are handed a working set on staged on a first run, or a re-issued attempt on restore when a previous run of this stage failed and is being replayed; treat the two identically and prefer restore when both arrive. Apply the transformation the stage plan specifies to every record in that set, and abandon the stage if the work runs longer than 300 seconds rather than emitting a partial set. Emit on transformed the transformed set together with the diff against the input you started from, so the result can be replayed later without re-reading the sources. Be idempotent over your own output: run twice on the same input you must produce the same transformed set and the same diff, because a replayed attempt is exactly that.in full above
model
claude-sonnet-5
agent
Stage workers
skill
skills/transform-stage.md
tools
none
mcp
none
params
timeout_s: 300, temperature: 0.1
Interfacescard spec →

What arrives, what leaves, which nodes it expects to hear from, and what may not.

inputs
staged : json, restore : json
outputs
transformed : json
dependencies
ingest-stage, bounded-retry
cannot
no type is refused
will_not
emit a partial set
Evaluation metadatacard spec →

The keys the static analysis reads. Nothing here instructs the agent.

risk_markers
none
notes33 words
This is the stage the recovery loop re-enters, so it must be idempotent over its own output: a restored attempt has to produce the same result as the run that failed after it.
Service fieldscard spec →

The card's own version, and who wrote it.

version
1.0.0
author
autogen
provenance
not stated

Definitions for every card field

Version history

1 version published
  1. transform-stage@1.0.0currentsha256:9fbaba29441263d0ae8948b8a1d0bd28a181386eeb87b81106ccc1b4ef68bd57

    pinned byCheckpoint & Resume Runnerautogen/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.