dsh-design-audit
DeepSeek Harness 的 UI/UX 审查与优化闭环:无头 Chrome 取证 → 确定性判据(WCAG 2.2 / 平台指南)→ 可粘贴 CSS 修复 → 复测差值验证
190 results
DeepSeek Harness 的 UI/UX 审查与优化闭环:无头 Chrome 取证 → 确定性判据(WCAG 2.2 / 平台指南)→ 可粘贴 CSS 修复 → 复测差值验证
Session image payload governor for DeepSeek Harness: audit which images a session still ships, pick the ones to keep, and move the rest out of the model context.
Strict check: verify code and commands against real checkers instead of reading them — Lean 4 kernel checking with an axiom audit, language type/syntax checks, and static defect rules.
Read-only update auditing for DeepSeek Harness profile plugins across npm, GitHub, and local package sources.
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.
Auditable, verifier-gated self-evolution for DeepSeek Harness.
判定 Python 代码是原创还是演绎作品:给出与上游最长相同表达片段、逐条豁免依据和风险清单。纯本地 AST 分析,零依赖、不联网、不调模型。
Permission matrix for DeepSeek Harness: 3 sandbox modes x 4 approval strategies = 9 selectable presets, with a global default, a separate LLM-robot default, four-tier risk policies, password-gated approval for high-risk operations, audit log and git checkpoint.
Audit a DSH profile for capabilities nothing consumes, and record completion claims with the evidence that supports them.
Audit a DeepSeek Harness session against the 14 consciousness indicator properties of Butlin, Long, Elmoznino, Bengio et al. (2023) and report an L1-L5 level
DeepSeek Harness security bundle: auto-audits newly installed plugins with the harness's own model (settings panel + batch tools), and ships the Security-Audit-Mode agent preset with the OpenAI Codex Security workflow skills and dsh_security_* scan tools. One package, one `dsh plugin add`.
Persistent, auditable uni-browser profiles for DeepSeek Harness.
Offline content-addressed proof for supplied DSH workflow duty separation
Layered, approval-gated, auditable cross-session memory for DeepSeek Harness — a capability seam (ctx.memory service + local SQLite provider + memory tool + frozen snapshot injection), not another memory warehouse
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.
Website cloning methodology skill for DeepSeek Harness: a real-source-first decision tree across static / React-Vue-Next / WebGL-Canvas sites, with CDP-driven recon, asset harvesting and pixel-diff audit scripts.
Deterministic fact producers for security auditing: which deleted code came from a security fix, and which changed symbols no test ever executed.
What will your agent's context cost before it runs? Audit instruction files, skills, MCP tool schemas and memory per source — with duplicates, always-on versus on-demand costs, and a budget you can enforce in CI.
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.
dsh plugin: shows the model name the upstream actually reported for each assistant step, backfilling pi-ai replay metadata from a non-invasive fetch observer when the adapter omits it
Isolated private chat with reversible context selection and retained audit records for DSH Web
DSH Cost Audit — per-turn and per-session token/CNY cost pills for the DeepSeek Harness Web GUI, including the compaction bill nothing else counts, plus a one-click advisor that re-measures whether its own advice actually saved money.