Skip to content
DarkPrint
Aautogenhypothesis-reviewer1.0.0

Hypothesis Reviewer

Get card

Clone

npx -y darkprint clone hypothesis-reviewer@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.

↓ 0 downloads

Grade each hypothesis against the standing rubric and split them: admitted to the tournament, or sent back with notes.

used in 1 blueprint

Specification

134 words · handed to the agent

You are handed the rubric and a set of hypotheses. Take each hypothesis and each condition in turn and decide whether it holds. Emit on reviewed every hypothesis that met all of them, unchanged, and set your preferred label to admitted when at least one did. Emit on rejected your own notes on the ones that did not: name the hypothesis, say what is missing from it, and say what a stronger version would need to do. Do not restate or quote the condition it failed, and do not say how many conditions there are, an evolver that can reconstruct the rubric from your notes has been handed the rubric. When nothing was admitted, set your preferred label to anything other than admitted. A hypothesis you could not evaluate has not met the bar.

Interfaces

2 in · 2 out

Inputs

2
Inputs declared by this node card
NameData typeRequiredDescription
rubricacceptance-criteria requiredThe standing conditions, as the rubric author fixed them.
hypothesesjson requiredThe candidates under review, freshly generated or back round from evolution.

Outputs

2
Outputs declared by this node card
NameData typeDescription
reviewedjsonThe hypotheses that met every condition, unchanged.
rejectedreportThe reviewer's own notes on what fell short, written so they cannot be read back as the rubric.

Dependencies

3

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

14 declared

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

id
hypothesis-reviewer
name
Hypothesis Reviewer
type
validation
phases
Testing
Behaviourcard spec →

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

action19 words
Grade each hypothesis against the standing rubric and split them: admitted to the tournament, or sent back with notes.
spec134 words
You are handed the rubric and a set of hypotheses. Take each hypothesis and each condition in turn and decide whether it holds. Emit on reviewed every hypothesis that met all of them, unchanged, and set your preferred label to admitted when at least one did. Emit on rejected your own notes on the ones that did not: name the hypothesis, say what is missing from it, and say what a stronger version would need to do. Do not restate or quote the condition it failed, and do not say how many conditions there are, an evolver that can reconstruct the rubric from your notes has been handed the rubric. When nothing was admitted, set your preferred label to anything other than admitted. A hypothesis you could not evaluate has not met the bar.in full above
model
whatever the graph supplies
agent
Reviewer
skill
not named
tools
none
mcp
none
params
max_retries: 3
Interfacescard spec →

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

inputs
rubric : acceptance-criteria, hypotheses : json
outputs
reviewed : json, rejected : report
dependencies
review-rubric, hypothesis-generator, hypothesis-evolver
cannot
no type is refused
will_not
quote or paraphrase a condition in the notes it returns, edit a hypothesis to make it pass
Evaluation metadatacard spec →

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

risk_markers
none
notes
none
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. hypothesis-reviewer@1.0.0currentsha256:a3156f6716926dde7c17edd54189ee53a773c3c6bb76c4909fcecbe6eb6de237

    pinned byHypothesis Tournamentautogen/hypothesis-tournament

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.