dsh-advisor
`dsh-advisor` is a DeepSeek Harness plugin ported from `rpiv-advisor`, designed to provide agents with a zero-argument `advisor()` tool that assists the primary model in decision-making. It forwards the full conversation history to a review model and returns either a plan or a stop signal. The plugin is disabled by default and must be enabled by configuring `provider` and `model`. It supports three levels of configuration through the plugin line, Web settings, and the `/advisor` command, allowing the primary model to invoke the tool on demand without altering the existing control flow.
Read Moredsh-advisor: An Independent Model Advisory Review Plugin for DeepSeek Harness
`dsh-advisor` is a plugin for DeepSeek Harness, designed to address issues such as agents performing destructive operations or deviating from instructions by using an independent review model. It provides independent model review, guidance injection, and zero-cost rule checking to ensure that agents can self-correct after executing tool steps. By default, the plugin uses the official DeepSeek API and supports configuring custom review models or regex rules to intercept dangerous operations. After installation, configure it in the Profile and restart to ensure permission security and improve workflow safety at a low cost.
Read Moredsh-advisor: Mounting Independent Review Sub-Models for DSH Sessions
dsh-advisor is a DSH plugin maintained by omdsh-dev that attaches an independent review model to each session, passively observes the main transcript, and injects suggestions categorized as nit, concern, or blocker after each stepped turn. The plugin is advisory-only and does not approve or reject actions on behalf of the main model. It supports a web Settings Advisor card and the `dsh-tui /advisor` command, and can be enabled by configuring the provider and model in `settings.yaml`.
Read MoreUsing dsh-advisor to configure the secondary model for passive review per round of DeepSeek Harness
dsh-advisor is a DeepSeek Harness plugin maintained by omdsh-dev. It attaches an independent review model to each session, injecting suggestions in the order of nit/concern/blocker, and does not approve actions or issue commands on behalf of the main agent. This article verifies the installation commands, configuration access control, Web/TUI usage and current MVP limitations based on the plugin directory page and GitHub README.
Read More