dsh-sandbox-permissions-not-strictly-wider-justification-empty-fix
修复 sandbox_permissions 和 justification 字段在 pwsh/bash/fs/dsh-sandbox 中的 no-op 升级报错。当会话已是 danger-full-access 模式时,模型携带空 justification 或重申同一模式被拒的问题。
72 results
修复 sandbox_permissions 和 justification 字段在 pwsh/bash/fs/dsh-sandbox 中的 no-op 升级报错。当会话已是 danger-full-access 模式时,模型携带空 justification 或重申同一模式被拒的问题。
Offline content-addressed proof for supplied vulnerability-remediation closure
Evidence-first inspection, compatibility verification, and quarantine tooling for DeepSeek Harness plugins.
Self-healing DeepSeek-Harness plugin: keeps confined pwsh/bash console windows invisible (SW_HIDE) on Windows — no black console flash on every sandboxed command.
Tolerate redundant sandbox_permissions requests: a tool call asking to escalate to a mode that is not strictly wider than the session's standing mode runs as-is at the standing mode instead of failing with "not strictly wider". Genuine escalations still route through the approval flow.
SSRF-resistant HTTP(S) fetch provider for DeepSeek Harness
HashiCorp Vault backend for the DeepSeek Harness credential seam: central secrets, AppRole machine auth, rotation without restart, and no long-lived provider key on the agent host
Normalize invalid non-widening sandbox escalation arguments before DSH tool dispatch.
Evidence-backed runtime nutrition labels for DeepSeek Harness plugins and tool namespaces
Cross-provider second-opinion advisor for dsh. v2 (simple-extensible-configurable-stable): DEFAULT ROUTE = official DeepSeek (deepseek-official/deepseek-v4-flash); model param optional everywhere; one-page visual settings panel; honest audit trail. Scout mode = official + live web_search. OpenAI-compatible endpoint + consult tool + DSH_MODEL rerouting.
Offline content-addressed proof that a temporary policy waiver stayed within approved bounds
Offline content-addressed evidence that an approved Windows control-plane change settled across required surfaces
Offline content-addressed proof of pseudonymous authority binding across DSH execution surfaces
Offline content-addressed proof for supplied credential-retirement closure
Blocks high-confidence secrets before dsh-TUI sends them to a model.
Bridges vendor agent SDKs into DeepSeek Harness as selectable LLM providers. Works in any DSH mode; DSH keeps its own prompt, tools, sandbox and approvals.
Offline DSSE/in-toto attestation evidence for DeepSeek Harness releases
DeepSeek Harness plugin: redacts known secret shapes in tool arguments and results behind stable placeholders, and appends preimage-free JSONL incidents
Read-only MCP server exposing the dsh-plugin-certification registry: certification grades, snapshots and five-dimension evidence for DeepSeek Harness plugins.
DeepSeek Harness plugin to audit and correct llm-pi-ai model capability declarations by probing the endpoint itself: out-of-range maxTokens, contextWindow, reasoning levels, image input.
Make a Git for Windows installation resolvable as `bash` inside the dsh host process, so the official bash shell stack (dsh-bash-sandbox + dsh-tool-bash) can run on Windows
Set up free, encrypted, deduplicated backups of remote Linux servers, pulled from a Windows PC with Restic — tar-over-SSH streaming, zero software installed server-side, with size verification against
Audit a DSH profile for capabilities nothing consumes, and record completion claims with the evidence that supports them.
DSH Metafolder Plugin — visual meta-folders for the DeepSeek Harness sidebar workspaces: nest real workspace folders under named, collapsible groups, by drag and drop or the row menu. Paths, permissions, sessions and workspace ordering are untouched.