dsh-risk-guard
Zero-interruption audit and fuse blocking for DeepSeek Harness: silently records every tool call with deterministic risk scoring, cumulative-risk bonuses, risk-level breakdowns and retention-based cleanup; blocks irreversible catastrophes (protected-path deletion, disk wipe, force-push to protected branches/refs, credential exfiltration), and renders a redacted /risk-guard operation bill with --since filtering.
零打扰审计与保险丝拦截：静默记录每次工具调用并做纯规则风险评分（含累积加分、风险等级分布、归档保留期清理），只拦不可逆灾难（受保护路径删除、磁盘擦除、force push 到受保护分支/引用、凭据外发），/risk-guard 输出脱敏操作账单并支持 --since 时间过滤。
How to install
dsh plugin add dsh-risk-guard About
dsh-risk-guard Zero-interruption audit + fuse blocking for DeepSeek Harness (dsh). Features **Silent audit** — records every tool call (tool name, redacted arguments, success/failure, risk tags, explainable score) into local JSONL files. No dialogs, no permission changes, no sandbox changes. **Deterministic risk scoring** — zero LLM calls, zero extra cost. Covers destructive deletes, credential reads, network egress, writes outside the workspace, dependency installs and heavy builds; rapid repeats and consecutive high-risk calls accumulate bonus score. **Insurance fuse** — blocks only irrevers…
Recommendation signals
Meta
- License
- MIT
- Language
- TypeScript
- GitHub stars
- 0
- mo. downloads
- 0
- Last push
- 2026-08-17
- Created
- 2026-08-17
Basic safety check
- Findings
- None
- Sources
- curated:awesome-dsh-plugin.com, curated:awesome-dsh-plugin/awesome-dsh-plugin
- Topics
- deepseek-harness, dsh-plugin, security
Related plugins
dsh-plugin-vetting
truelove-dreamer/dsh-plugin-vetting
Vets third-party plugins before you trust them: static scan for malicious patterns (exfiltration, credential access, obfuscation, persistence) and over-privileged path use, transitive-dependency coverage, official-package hash baseline for supply-chain tamper detection, and an optional plugin-tool call gate.
dsh-approval-llm
Letter2025/dsh-approval-llm
Model-based permission approval: an approval-request answerer backed by a separate reviewer model.
dsh-auto-review
PerryLink/dsh-auto-review
Second-model auto-review on the approval answerer chain: a read-only reviewer subagent returns structured allow/deny verdicts with reasons, fail-closed by default.
dsh-auto-classifier
PAKIKNOWLEDGE/dsh-auto-classifier
Autonomous permission classifier for the auto preset: tool-scoped allow/deny rules, an LLM semantic judge, and git checkpointing for unattended sessions.