maiziman/deepseek-harness-portable--plugins-dsh-model-capabilities ↗★ 1
@maiziman/dsh-model-capabilities
Automatic reasoning and image-input capability discovery for custom DeepSeek Harness model providers
安装
$
npx -p @deepseek-ai/dsh dsh plugin --profile web add github:maiziman/deepseek-harness-portable#b8ebad37978590c0c55ea2f041a10338b1a33b8e&path:plugins/dsh-model-capabilities说明文档
阅读完整 README ↗Configuration
The shipped Bundle uses:
- insert:
- id: model-capabilities
name: '@maiziman/dsh-model-capabilities'
config:
metadataDiscovery: true
activeProbePolicy: local-only
reasoningProbe: true
visionProbe: true
reasoningProbeEfforts: [low, high, max]
probeConcurrency: 1
requestTimeoutMs: 10000
probeMaxTokens: 128
| Field | Meaning |
|---|---|
metadataDiscovery | Read non-generating model metadata. |
activeProbePolicy | never, local-only, or always; local-only recognizes exact localhost plus literal loopback, link-local, RFC 1918, and unique-local IPv6 addresses. DNS names require always. |
reasoningProbe | Allow reasoning requests when the active-probe policy permits them. |
visionProbe | Allow the inline image request when the active-probe policy permits it. |
reasoningProbeEfforts | Effort spellings tested independently; only positively observed levels are saved. |
probeConcurrency | Maximum simultaneous reasoning probes for one model. |
requestTimeoutMs | Bound for each metadata or active request. |
probeMaxTokens | Maximum output tokens requested by a probe. |
To override the defaults, target the model-capabilities row in the profile's own cordis.patch.yml. A later row configuration replaces the complete earlier configuration, so restate every field you want to keep.