dsh-security-requirements
SQL 优化模式:EXPLAIN 分析、索引策略、N+1 解决、查询改写。受 wshobson/agents(38k★ MIT)启发。
173 results
SQL 优化模式:EXPLAIN 分析、索引策略、N+1 解决、查询改写。受 wshobson/agents(38k★ MIT)启发。
DSH++ — a local-first control plane for DeepSeek Harness. Plugin security auditing and workflow regression testing for dsh, in a CLI plus a loopback web console.
Podman-backed execution for DeepSeek Harness (dsh)
SQL 优化模式:EXPLAIN 分析、索引策略、N+1 解决、查询改写。受 wshobson/agents(38k★ MIT)启发。
Indirect prompt-injection guard for DeepSeek Harness: taints tool output by origin and gates privileged tool calls that follow untrusted content
Run an agent unattended for hours and get a morning brief that is an audit, not a summary. Signs a manifest before the work, holds the run to it, and reports the difference.
Deny an AI agent access to .env files, credential stores, keys and any path you hide — across file tools, shell commands, search selectors and run_code. A DeepSeek Harness (dsh) plugin.
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.
Fail-closed secret-like payload guard for DeepSeek Harness tool calls.
DSH (DeepSeek Harness) plugin that bundles the battle-tested 'safe plugin installation' skill: audit candidates, install one at a time, verify through five levels, and uninstall cleanly. Built from a real beginner's painful install journey.
Verification toolkit for DeepSeek Harness agents: evidence-based claim checking against workspace files with line citations, config validation (JSON/YAML), HTTP URL status checks, npm package checks, GitHub repo submission-readiness checks, and batch verification with bounded concurrency and per-result confidence grading
Offline DSSE/in-toto attestation evidence for DeepSeek Harness releases
Static trust audit for DeepSeek Harness plugins: capabilities, injected prompts, token cost, provenance, and a reproducible 0-100 trust score
Offline content-addressed proof that a temporary policy waiver stayed within approved bounds
Plugin security inspector for the DeepSeek Harness web GUI: statically audits installed plugins (dangerous API patterns + dependency review), layers an AI (default-model) online audit, and renders a green/yellow/red report panel.
Audit a DSH profile for capabilities nothing consumes, and record completion claims with the evidence that supports them.
OpenKapsel workspace bridge for the DeepSeek Harness: operate a remote OpenKapsel workspace (files, shell, context, memory, sharing, preview) through skill-driven tool calls using only its workspace URL and control token.
Human intent verification and cryptographic action authorization for AI agents. AI can propose an action. Only a human can authorize it.
macOS Keychain credentials provider for DeepSeek Harness (dsh) — refs and records both stored in the login keychain, never in a plaintext file
dsh-voredteam —— 单模式(网络安全模式)DSH 插件合集:黑板书 Fact/Intent/Hint 引擎 + 总控派发专业子 agent + 作战面板/设置 两个面板 + 唯一门禁(禁 DDoS/禁爆破/模糊测试低频)。
DSH memory plugin gated by the TypeSafe Jev decision model: mem_* tools whose every read/write is a typed judgement (choice/noul) — write gate, recall gate and a fail-closed injection gate with budget, audit log and restore/merge.
AI-reviewed auto-approval for DeepSeek Harness: judges each pending approval ask with one auxiliary model call before any human answerer runs, auto-grants the safe ones, and escalates everything else with the review analysis attached
Codex-style auto-approval for DeepSeek Harness. An independent reviewer decides allow/deny, fail-closed, with a rationale for every decision. Engine: an LLM route, or TypeSafe's Jev (`reviewer.engine`). · DSH 插件:独立复核器裁决,引擎可选 LLM 路由或 Jev。
Explain what a pending dsh approval will actually do, with one LLM call: an Explain button beside Allow/Reject plus the Host /explain command behind it