Weighted Vote

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

id

weighted-vote

name

Weighted Vote

type

decision

phases

Implementation

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

action27 words

Score every competing proposal under the weighting rule and emit the winner together with the full tally, so the choice is auditable rather than a black box.

spec97 words

Score each of the competing proposals you are handed, weighting every one by the reputation of the validator behind it, or by the scoring rubric supplied for this run, when there is one, which replaces the default weighting entirely. Emit the highest scorer on chosen, tagged with the margin it won by, and settle an exact tie in favour of the more conservative proposal rather than picking arbitrarily. Emit every proposal with its score and the weights that produced it on tally: the disagreement is the part worth reading, so never report the winner on its own.in full above

model

whatever the graph supplies

agent

Consensus judge

skill

skills/weighted-vote.md

tools

none

mcp

none

params

weighting: validator-reputation, tie_breaker: conservative-proposal

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

inputs

proposals : json, rubric : text

outputs

chosen : json, tally : json

dependencies

conservative-solver, exploratory-solver, bounded-retry

cannot

no type is refused

will_not

report the winner without the tally

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

risk_markers

none

notes35 words

The tally is the point. A vote that only emits the winner is a black box, and the whole reason this node exists instead of a coin flip is that the disagreement is worth reading.

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

version

1.0.0

author

autogen

provenance

not stated