DSHUNIVERSE
Back to catalog

GITHUB TOPIC

ai-safety

8projects include this topic

The "ai-safety" topic on GitHub groups 8 open-source projects in the DeepSeek Harness (DSH) ecosystem, led by dsh-auto-review with 198 GitHub stars. dsh-auto-review — Second-model AI auto-review for DeepSeek Harness approval requests: a read-only reviewer subagent returns structured allow/deny verdicts with reasons, fail-closed by default, fully auditable from the session log (approval/asked -> autoReview/verdict -> approval/decided). Every project here is indexed by DSH Universe with live GitHub data — stars, activity and install status — so you can compare and install directly.

{count} projects

Exact GitHub Topic match

Security & governancePlugin

Second-model AI auto-review for DeepSeek Harness approval requests: a read-only reviewer subagent returns structured allow/deny verdicts with reasons, fail-closed by default, fully auditable from the session log (approval/asked -> autoReview/verdict -> approval/decided).

Structure check pending
ai-safetyapprovalauto-reviewcordis
198 2 0
8 days
Security & governancePlugin

Claude Code-style declarative permission rules for DeepSeek Harness: ordered allow/deny/ask rules with tool-name, argument (glob/regex), and workspace-path matching on the tools/pre-execute waterfall, session-log audit, and HMR reload.

Structure check pending
ai-safetyallow-deny-askapprovalcordis
114 3 4
8 days
Development toolsSkill

sofagent

KongFangXun

sofagent is an open-source FDE Harness for AI coding agents — on entry, write your business judgment into files (workflows, ontology, AI nodes); after departure, audit every change against them. 25 codified audit rules, tamper-evident chain, snapshots, and an MCP toolset (104 tools, 11 plugins). Everything can be FDEing.

Structure check pending
agent-auditagent-governanceai-agentsai-safety
51 9 NaNm
1 days
Agents & sessionsPlugin

Execution-time drift firewall for long-running DeepSeek Harness agents. Real-Harness tests: unsafe stale mutations 12/12 native -> 0/12; valid controls 7/7 both; post-SIGKILL unsafe continuation 2/2 -> 0/2.

Structure check pending
agent-harnessagent-orchestrationagent-planningagent-safety
3 0 0
1 months
Security & governancePlugin

Deterministic fail-closed tool-call authorization for DSH with evidence: allow/block/ask policy gate plus approval-chain deferral.

Structure check pending
agent-safetyagent-toolingai-safetydeepseek-harness
3 0 0
1 months
Models & MCPPlugin

DeepSeek Harness development forge: approval guard, dev Skills, GitHub/browser capabilities and Token Watch consumption supervision — install and get to work.

Structure check pending
ai-agentsai-safetycordisdeepseek-harness
2 0 0
1 months
DSH PluginsPlugin

⚡ Millisecond System-1 judgement for every tool call in DeepSeek Harness — Jev-powered risk classification & evidence-gated auto-approval. Fail-closed by construction. dsh 生态第一个 System-1 决策插件

Structure check pending
agent-harnessagentic-aiai-agentsai-safety
1 0 NaNm
10 days
Security & governancePlugin

Make the DSH Coding workflow more deterministic! Delivered as both a DSH preset and a global plugin (usable in any session). Converge → normalize → implement: parallel TDD dual-loop convergence gates, adversarial cross-source review, per-source citation verification, advance technical collision search — every gate keeps an honest record, never silently turning green

Structure check pending
agent-reliabilityai-safetycoding-agentsdeepseek-harness
1 0 0
1 months