Get card
Download
proximity-clusterer@1.0.0.yamlClone
npx -y darkprint clone proximity-clusterer@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.
Group the ranked hypotheses by how close they are to one another, so the spread of the field is visible rather than just its top.
used in 1 blueprint
Specification
97 words · handed to the agentEmbed each hypothesis and group them by proximity. Emit the groups on clustered, each carrying its members unchanged and in the order they arrived, so the ranking survives the grouping. A group of one is a real result and must be emitted as such: an isolated hypothesis is the most interesting thing this node can find, and folding it into the nearest group to tidy the output destroys exactly that. Do not rewrite, summarise or retitle a hypothesis to make a grouping cleaner, and do not write anything from this run back into the store you search.
Interfaces
1 in · 1 outInputs
1| Name | Data type | Required | Description |
|---|---|---|---|
| ranked | json | required | The ordered hypotheses with their comparison records. |
Outputs
1| Name | Data type | Description |
|---|---|---|
| clustered | json | The hypotheses grouped by proximity, ranking preserved inside each group. |
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
15 declaredWho the node is. The id is the key the DOT pins.
- id
- proximity-clusterer
- name
- Proximity Clusterer
- type
- tool
- phases
- none declared
What it does, and the prose the agent is handed when the graph runs.
- action25 words
- Group the ranked hypotheses by how close they are to one another, so the spread of the field is visible rather than just its top.
- spec97 words
- Embed each hypothesis and group them by proximity. Emit the groups on
clustered, each carrying its members unchanged and in the order they arrived, so the ranking survives the grouping. A group of one is a real result and must be emitted as such: an isolated hypothesis is the most interesting thing this node can find, and folding it into the nearest group to tidy the output destroys exactly that. Do not rewrite, summarise or retitle a hypothesis to make a grouping cleaner, and do not write anything from this run back into the store you search.in full above - model
- whatever the graph supplies
- agent
- Clusterer
- skill
- not named
- tools
- Vector store
- mcp
- qdrant
- params
- min_cluster_size: 1
What arrives, what leaves, which nodes it expects to hear from, and what may not.
- inputs
- ranked : json
- outputs
- clustered : json
- dependencies
- elo-tournament
- cannot
- no type is refused
- will_not
- fold a single-member group into its nearest neighbour, write this run's own text back into the store
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- proximity-clusterer@1.0.0currentsha256:c5497b36dc35154f14ac2b4ed43af048181b11ceb53639dfea3076fbe77e9908
pinned byHypothesis Tournament
autogen/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.