dsh-permission-matrix
Matrix-based permission and approval plugin for DeepSeek Harness with 9 presets, risk classification, and password-gated approvals.
Install
$ dsh plugin --profile <profile> add dsh-permission-matrixPlugin Overview & Capabilities
AI-assisted organization based on the public repository snapshot. The content must be grounded in source evidence and does not replace compatibility or security verification.
Key Capabilities
- Provides 9 configurable permission presets combining 3 sandbox boundary levels and 4 approval strategies.
- Implements separate default presets for global Web sessions and workspace-based LLM robot sessions.
- Features a 4-tier risk classification engine (Low, Medium, High, Hard) with execution surface parsing and LLM judge fallback.
- Supports password-gated approval overlay and token generation for high-risk and sensitive operations.
- Performs automatic pre-execution git snapshots and writes detailed JSONL audit logs.
Useful For
- Protecting host environments against destructive bash or PowerShell operations during autonomous agent sessions.
- Assigning distinct security profiles for automated chat robots (e.g. QQ, WeChat, Telegram) versus interactive web users.
- Requiring password confirmation before executing critical filesystem or git operations.
Who It Fits
- DeepSeek Harness users running autonomous agents on local development machines.
- Teams integrating chat bots and headless automation requiring strict permission sandboxes.
- Security-focused developers needing granular audit trails and password gates for high-risk operations.
Documented Limitations
- Conflicts with dsh-auto-classifier if both attempt to patch the permission preset line simultaneously.
- Multi-level obfuscated code concatenation or dynamic scriptblock execution may bypass static rule analysis and rely on the sandbox or LLM judge fallback.
- Operations configured for password approval require a password to be set; otherwise, they fail closed and are denied.
DSH Compatibility
Version-specific runtime evidence collected by DSH Plugin. A missing result means we have not tested that combination yet.
Security Signals
Objective signals discovered from package metadata and source inspection. These are not a guarantee that a plugin is safe.
package.json declares 0 runtime, 0 development, 2 peer, and 0 optional dependencies.
package.json declares DSH bundle metadata.
GitHub reports the repository license as MIT.
A root package.json was captured and can be inspected by the registry.
Public GitHub source metadata is available for this registry snapshot.
Source & Registry Notes
Public provenance, Registry classification, and the latest source check for this entry, kept separate from runtime verification.
- Source repository
- zhang8019/dsh-permission-matrix
- Registry source
- GitHub · dsh-plugin topic
- Registry classification
- Plugin
- Source checked
- f4eab4b · 2026-09-13
This project is independently indexed from public source information. DSH Plugin is not affiliated with DeepSeek or the plugin author. Always check the author repository before installation.
Repository Activity
- GitHub stars
- 0GitHub stars
- Forks
- 0
- Open issues
- 0
- Last commit
- 2026-09-11
- Last release
- 2026-09-11
Related DSH Plugins
Ranked by overlapping capabilities, use cases, plugin type, categories, and DSH profile.
Pre-execution safety net blocking destructive shell commands and secret leaks in coding agents.
View plugin →Second-model AI auto-review for DeepSeek Harness approval requests with fail-closed security.
View plugin →Agent-native multi-market trading terminal and quantitative research plugin bundle for DeepSeek Harness.
View plugin →Security red-teaming and prompt rule management plugin for DeepSeek Harness web and desktop profiles.
View plugin →