consensus-publish-guard

INLINECODE0 protects public-facing content before release.

What this skill does

• - reviews content drafts with persona consensus
• flags policy/legal/sensitive-risk patterns
• decides INLINECODE1
• emits rewrite patch when fixable
• persists decision artifacts to board state

Why this matters

Public content creates brand, legal, and trust exposure. Consensus review creates a safer publish gate than single-pass generation.

Ecosystem role

Uses deterministic logic from consensus-guard-core and existing persona inputs, persisted via consensus board primitives.

Good fit for

• - AI-assisted social/media pipelines
• product launch copy checks
• policy-sensitive communications

Runtime, credentials, and network behavior

• - runtime binaries: node, INLINECODE4
• network calls: none in the guard decision path itself
• environment config read by this package: CONSENSUS_STATE_FILE, INLINECODE6
• filesystem writes: board/state artifacts under the configured consensus state path

Dependency trust model

• - consensus-guard-core is the first-party consensus package used in guard execution
• versions are semver-pinned in package.json for reproducible installs
• this skill does not request host-wide privileges and does not mutate other skills

Install assumptions

This repository currently expects a local sibling checkout of consensus-guard-core.

CODEBLOCK0

Then install dependencies in this repo:

CODEBLOCK1

Install (registry)

CODEBLOCK2

Quick start

CODEBLOCK3

Tool-call integration

This skill is wired to the consensus-interact contract boundary (via shared consensus-guard-core wrappers where applicable):

• - readBoardPolicy
• getLatestPersonaSet / getPersonaSet
• writeArtifact / writeDecision
• idempotent decision lookup

This keeps board orchestration standardized across skills.

Invoke Contract

This skill exposes a canonical entrypoint:

• - INLINECODE10

INLINECODE11 starts the guard flow and executes deterministic policy evaluation with board operations via shared guard-core wrappers.

external_agent mode

Guards support two modes:

• - mode="external_agent": caller supplies external_votes[] from agents/humans/models for deterministic aggregation.
• INLINECODE14: requires an existing persona_set_id; guard will not generate persona sets internally.