dsh-skill-explorer
zhu1090093659
Skill center for the dsh web GUI: browse all loaded skills grouped by source, enable or disable model invocation, create new skills, and delete into a recoverable trash.
Installed cleanly when we ran it
dhicoc/dsh-reverse-skill
Complete reverse-skill pack (85 SKILL.md) as a DeepSeek Harness Cordis plugin: reverse engineering, authorized pentesting and security-research skill router.
Install
via npm
Resolves a published tarball rather than cloning the repository, and installs without any extra setup. Swap `web` for your profile name if you run another one.
via GitHub · npm package
Installing from GitHub runs the project's build script, which pnpm blocks until you allowlist it — run the command once and pnpm prints the exact key to add under `allowBuilds` in ~/.dsh/profiles/web/pnpm-workspace.yaml.
The desktop app keeps its own profile, desktop, and only the dsh command it ships may change it — the npm-installed dsh refuses that profile. Install the command from the app menu (Manage dsh Command…), open Desktop once, quit it fully, run the line below, then reopen Desktop.
via npm
Resolves a published tarball rather than cloning the repository.
via GitHub · npm package
Installing from GitHub runs the project's build script, which pnpm blocks until you allowlist it — run the command once and pnpm prints the exact key to add under `allowBuilds` in ~/.dsh/profiles/desktop/pnpm-workspace.yaml.
Our install check ran in a web profile on npm's dsh, not in the desktop app.
What happened when we ran it
DeepSeek Harness 0.2.2-rc.5 refuses the latest release
From 0.2 DeepSeek Harness checks a plugin's dsh range before installing anything, and refuses the whole command when the running version is outside it. The latest release asks for ^0.0.1-rc.1; we ran 0.2.2-rc.5.
No release of it accepts DeepSeek Harness 0.2.2-rc.5 yet. `dsh plugin allow-version … --accept-risk` forces one in, at the risk the harness itself warns about; otherwise wait for an update.
Why DeepSeek Harness refuses plugins, and every way around it ›
Every command here is run in a throwaway container against a clean profile, and the result is whatever the harness recorded — not a guess from the source. Last run yesterdayon DeepSeek Harness 0.2.2-rc.5.
Due diligence
Security
Not measured — DSH refused the latest release before running anything.
Source: our sandbox install on DeepSeek Harness 0.2.2-rc.5, yesterday.
package.json declares no preinstall, install or postinstall script.
Source: package.json on GitHub, read at our last sync. prepare is not recorded; it runs only on a GitHub install, and shows up under build scripts when pnpm blocks it.
shell commands (bash, exec, spawn …)
Source: a keyword match on the README — a hint about what the plugin touches, not an analysis of its code.
Not an audit. These are the record of one install and two text scans; nobody has read this plugin's code on your behalf. Plugins run with your agent's permissions.
What it is — This is a DeepSeek Harness plugin that wraps the reverse-skill pack into a Cordis plugin providing reverse engineering, authorized pentesting and security-research skill routing.
Who it is for — If you are performing authorized pentesting and security research with DeepSeek Harness, you can install it to get these skills. It will automatically register 85 SKILL.md files into ctx.skills. You do not need to install it if you only use preset tools without needing custom skill routing.
Watch out — Sandbox testing shows DeepSeek Harness 0.2.2-rc.1 refuses to install the latest version because its declared dsh version range does not include 0.2.2-rc.1. No version of it currently accepts this plugin. It executes shell commands and no other issues were observed.
The verdict — The current version range does not match, so I would wait for the author to update it before considering an install because it provides these skill routes.
Generated by grok-4.6, and a starting point rather than a verdict. Where it says a plugin installs or does not, that is from a real run in a clean profile — everything else is read off the repository. Trust the source over this.

dsh-reverse-skill is a DeepSeek Harness plugin that registers 85 reverse-engineering, authorised penetration-testing, security-research and CTF skills as a Cordis skill provider. DeepSeek Harness discovers the packaged SKILL.md files under skills/ and CTF-Sandbox-Orchestrator/: src/index.ts recursively scans both trees, normalises front matter, builds SkillCandidate records and exposes their full bodies through ctx.skills. The model can retrieve them through ctx.skills and tool-skill, while users can invoke eligible skills by name.
The package is intended for authorised reverse engineering, penetration testing and security research. Its terminal surface is relevant because these skills can guide terminal-oriented security workflows; authorisation for the target system remains the operator’s responsibility. The plugin is a poor fit where only a small subset of these skills is needed, where existing local skills already cover the same material, or where external tooling must be available automatically: referenced MCP servers such as burp-mcp require separate configuration.
DeepSeek Harness does not include the upstream agents/*.yaml definitions, because those OpenAI Agents SDK agents cannot be mapped to its supported ctx.subagent targets. It also does not enforce allowed-tools or disallowed-tools; those constraints must be handled at the harness layer.
DeepSeek Harness loads the package through its dsh.bundle manifest and inserts the reverse-skill Cordis plugin into the active profile. A package reference can also be placed in the plugins list:
plugins:
- "@dhicoc/dsh-reverse-skill"
As a preset fallback, DeepSeek Harness accepts these paths under skills.local.customSkillDirs:
skills:
local:
customSkillDirs:
- "./dsh-reverse-skill/skills"
- "./dsh-reverse-skill/CTF-Sandbox-Orchestrator"
The documented discovery order is project .dsh, project .agents, customSkillDirs, user .dsh, then user .agents. Discovery is flat, so nested skills must be exposed through directories containing SKILL.md.
Run npm run build to compile TypeScript into lib/ and lib/types/. Run npm test to rebuild the plugin and test the registered provider. The test checks that all 85 skills are listed, names are unique, and each skill returns a non-empty body through get(). It also checks SKILL.md files containing a UTF-8 BOM and CRLF line endings.
During apply(ctx), DeepSeek Harness calls ctx.skills.registerProvider(...). The provider recursively scans skills/ and CTF-Sandbox-Orchestrator/, parses front matter, promotes metadata.user-invocable to user-invocable, converts when_to_use to whenToUse, and stores each resource with its directory path. The complete document body is returned when a skill is requested. Adding or removing a SKILL.md changes the discovered set without editing a candidate list.
The upstream agents/*.yaml files are not included. DeepSeek Harness does not enforce allowed-tools or disallowed-tools. MCP references in skill content, including burp-mcp, require separate entries in mcp.servers. The package is documented for authorised use only.
Written from the project's own documentation and kept in sync with it. Where the two disagree, the source is authoritative — read the README on GitHub
Same category
zhu1090093659
Skill center for the dsh web GUI: browse all loaded skills grouped by source, enable or disable model invocation, create new skills, and delete into a recoverable trash.
Installed cleanly when we ran it
QingYunA
Answers a question with a one-page visual HTML explainer: the model writes a short Markdown draft and the bundled `am` CLI builds the page, with flow, sequence, tree, timeline and ER diagrams, tables and callouts.
Installed cleanly when we ran it
iflytek
Browse and search a SkillHub registry and manage local skill installation, updates, and removal inside DeepSeek Harness. Targets DSH 0.2.0-rc.2 and 0.2.1-rc.2; not yet compatible with 0.2.1-rc.3.
No one-line install — this plugin lives inside a larger repository and publishes no npm package.
chuspeeism
现代化可灵活嵌入的任务面板,支持 Codex、DeepSeek Harness
Installed cleanly when we ran it
Alisa0808
Turn one topic into a finished Vox-style paper-collage explainer/ad video — automated end to end on Atlas Cloud + ffmpeg. An agent skill.
Installed cleanly when we ran it
zhuyansen
Daily radar for new AI terms: names of new models, products and repos, caught on X, Hugging Face, GitHub and sitemaps before keyword tools list them.
Installed cleanly when we ran it