openbkn-ai/bkn-dsh

C3

OpenBKN's DeepSeek Harness Web UI plugin enables business users to safely, accurately and reliably utilize the enterprise's business knowledge network to complete business analysis and decision-making

★ 10+ · openbkn-ai/bkn-dsh source on GitHub · this plugin in the registry

openbkn-ai/bkn-dsh is a DeepSeek Harness plugin rated C3 — powerful capability combined with sensitive behavior. It patches the dsh runtime, can rewrite the system prompt, can spawn subprocesses, executes system commands.

Installable plugin — declares a dsh.bundle manifest

What it can do

CapabilityFlagEvidence
patches the dsh runtimeruntime_patch./cordis.patch.yml
can rewrite the system promptprompt_surfacesubprocess, systemPrompt
can spawn subprocessessubprocess_serviceinject: subprocess
executes system commandsexec×45 in authored code, e.g. runtime/prepare-compatible-runtime.mjs:3, runtime/prepare-compatible-runtime.mjs:3
decodes base64 payloadsbase64_decode×1 in authored code, e.g. packages/openbkn-business-context/src/platform-csv-export.ts:133
starts a network servernet_server×2 in authored code, e.g. scripts/probe-openbkn-sdk021-contract.mjs:161, scripts/probe-openbkn-sdk021-contract.mjs:199
powerful capability in test/example code onlydev_surfacein code that does not ship: system-prompt/assemble

Tool registrations

1 confirmed registration point(s); analysis is incomplete.

Services it injects

agents credentials loader locale openbknWorkspaceBindingRegistry remote slots storageDomain subprocess systemPrompt tools

Hooks it attaches

agent/created agent/pre-step

Outbound domains

openbkn.example.com www.npmjs.com

Environment variables it reads

SystemRoot BKN_CONFIG_DIR SDK_CONTRACT_ORIGIN PROGRAMFILES LOCALAPPDATA OPENBKN_PLATFORM_PIN_FILE

How to read this

Levels measure capability surface and transparency, not maliciousness. A C3 plugin can be entirely legitimate — a desktop shell genuinely needs subprocesses. The point is that you can see this before installing. See the levels explained and how dsh plugins work.

Findings come from static analysis of shipped code; nothing is executed. Think a flag is wrong? Open an issue — every flag cites the file and line it came from.