979569650/dsh-typesafe ↗★ 1

dsh-typesafe

提供类型化决策模型、置信度路由及提示词注入防御。 适合需要高可靠性类型化决策、成本监控和安全防护的Agent工作流。

套件
dsh-typesafe
相容性
待驗證
Harness 依賴範圍
^0.1.5-rc.2
Cordis 依賴範圍
^4.0.2
版本
0.1.0
授權
MIT
最近更新
2026年9月22日

安裝

$npx -p @deepseek-ai/dsh dsh plugin --profile web add github:979569650/dsh-typesafe

Configuration reference

Every key is defaulted, so id + name alone is a working install. Edit them in the settings card, or override the row in a profile cordis.patch.yml.

KeyDefaultMeaning
enabledtrueMaster switch. Off removes the prompt section and the guard.
apiKeyEnvTYPESAFE_API_KEYCredential reference to resolve.
baseURLhttps://api.typesafe.aiAPI origin.
modeljev-latestAlias or pinned version id. Pin a version if you tuned thresholds against it.
timeoutMs30000Per-request timeout.
maxStateChars60000Rejects oversized calls before they are billed.
inputPricePerMTok0.042Cost reporting only.
promptEnabledtrueWhether to tell the agent when to use the tools.
promptOrder700Prompt section placement, in the behavioural-rule band (TEAM_POLICY 600 … PTC_ONLY 800).
guardEnabledtrueScreen untrusted tool results.
guardToolsweb_fetch, web_search, read_page, read_url, browse, scrape, mcp__*__fetch, mcp__*__fetch_content, *fetch_contentWhich tool results to screen. Globs, so one entry covers a family. Only tools that bring outside text in belong here.
guardWarnThreshold0.5Injection probability at which a result is flagged.
guardMinChars200Skip shorter results.
guardMaxChars6000Chunk size for screening. Every chunk is screened, so this trades request count against request size — it is not an excerpt budget.