dsh.so
EN
English中文
@wasteoneslife/dsh-skill-creator avatar

@wasteoneslife/dsh-skill-creator

DeepSeek Harness plugin: create spec-compliant skills from the Web UI (Huawei Cloud / generic / REST API), with a 10-point evaluation loop

PluginDeveloperBrowserNetworkCreator & socialWeb search
Verification
L5 · Run tested
Verified
Risk
Critical
Health
Active
Trust
Silver

Install verification: runtime verified (L5) · tested on dsh 0.2.1-rc.2 · last checked Install verification details →

Install verification badge dsh 0.2.1-rc.2

What it doesAI

Creates spec-compliant DSH skills from the Web UI (Huawei Cloud, generic, REST API) with a 10-point evaluation loop.

  • Creates spec-compliant skills directly from the Web UI
  • Supports Huawei Cloud, generic and REST API skill types
  • Built-in 10-point evaluation loop to validate skills
developerAIcreatorAI

Capability tags are AI-inferred or rule-extracted from the README (see badge) — vocabulary-limited, not install-verified. AI = inferred; README = literal keyword from the readme.

Installation

dsh plugin --profile web add @cloudplugin/dsh-skill-creator
verified 0.0.4

Install method: npm · runtime verified (L5)

Verification & Compatibility

Grouped by plugin version: each card lists the dsh versions actually tested against that artifact version and their install verdicts (L5 wins). Untested combinations are omitted — absence means untested, never assumed compatible. L1–L3 are a plugin-level static archive (independent of the dsh version — the same verdict on every dsh row): they sit in each row's level grid next to L4/L5, marked "static", while their evidence is stored once per plugin inside that row's "checks & evidence" instead of being duplicated per version.

External credentials (plugin-level): none detected

Declared DSH compatibilityfrom package.json · peerDependencies
author declared no dsh version requirement

○ The author declares no dsh version requirement — install is not blocked by version.

Install compatibility matrix (rows = plugin versions〔latest 5〕, columns = dsh versions〔latest 5〕)
At-a-glance → evidence in the cards below
plugin ＼ dshdsh 0.2.1-rc.2
current
dsh 0.2.1-alpha.4
0.0.4✓·
Install compatibility (by plugin version, dsh versions within)
Legend:✓ L5 runtime verified○ ecosystem plugin◼ ecosystem app✗ failed
0.0.4npm✓ 1 runtime-verified2026-10-19

@cloudplugin/dsh-skill-creator

✓dsh 0.2.1-rc.2current2026-10-19L5 · runtime verified
Environment & artifact
  • • Node.js: v24.14.0
  • • DSH: 0.2.1-rc.2
  • • Plugin artifact: npm · @cloudplugin/dsh-skill-creator@0.0.4
Levels (L1–L3 plugin-level static archive · L4/L5 newest real verdicts for this combination)
L1 static
Passed
L2 static
Passed
L3 static
Passed
L4 sandbox
Passed
L5 runtime
Passed
Install + runtime smoke verified in the dsh sandbox — installs and runs
L5 · runtime verified · valid for dsh 0.2.1-rc.2

ID: dsh-skill-creator-5@0.0.4@dsh0.2.1-rc.2 · profile: l4-sandbox

Issued: 2026-10-19 · expires: 2026-10-26

View checks & evidence
L1 · Found✓ Passed
Defaulted from L3✓ Passed
defaulted to passed — L3 install spec passed (full-registry L3 sweep policy)
L2 · Structured✓ Passed
Defaulted from L3✓ Passed
defaulted to passed — L3 install spec passed (full-registry L3 sweep policy)
L3 · Install spec✓ Passed
Install command parsed✓ Passed
Source: registry verification · Method: install spec parse · Captured 2026-10-18
DSH version declared✓ Passed
Source: registry verification · Method: DSH version declaration · Captured 2026-10-18
L4 · Install tested✓ Passed
Install executed in confined sandbox✓ Passed
Source: Windows ACL sandbox · Method: sandbox install run · Captured 2026-10-19
spec @cloudplugin/dsh-skill-creator · channel npm package · outcome ran · exit 0
check took: 2.4s
Install succeeded✓ Passed
Source: Windows ACL sandbox · Method: sandbox install run · Captured 2026-10-19
spec @cloudplugin/dsh-skill-creator · channel npm package · outcome ran · exit 0
check took: 2.4s
Installed artifact confirmed✓ Passed
Source: Windows ACL sandbox · Method: sandbox install run · Captured 2026-10-19
spec @cloudplugin/dsh-skill-creator · channel npm package · outcome ran · exit 0
check took: 2.4s
L5 · Run tested✓ Passed

active in loader tree (150 entries)

Sandbox install passed✓ Passed
Source: Windows ACL sandbox · Method: sandbox install run · Captured 2026-10-19
spec @cloudplugin/dsh-skill-creator · install mode reused L4 scratch
check took: 65ms
Web boot ready✓ Passed
Source: DSH runtime · Method: web boot · Captured 2026-10-19
ready line dsh web: http://127.0.0.1:63883/?token=…
HTTP endpoint served✓ Passed
Source: DSH runtime · Method: HTTP request · Captured 2026-10-19
path / · HTTP 200 · took 65ms
Plugin active in inventory✓ Passed
Source: DSH runtime · Method: plugin inventory query · Captured 2026-10-19
verdict active · entries 150
match: include:dsh-skill-creator · @cloudplugin/dsh-skill-creator · phase active
phases: active 122 · null 28

L1–L3 are plugin-level static conclusions (baseline, identical across dsh versions): evidence is stored once per plugin.

Only the latest 3 published versions are shown; source-snapshot (commit hash) groups, unstated groups and older versions are omitted — the full archive stays in the data layer (snapshot/unstated groups appear only when a plugin has no published version).

Author badges

Embed the official badges in your README to showcase security & install-test status:

dsh.so risk
`[![dsh.so risk](https://www.dsh.so/badge/dsh-skill-creator-5.svg)](https://www.dsh.so/artifact/dsh-skill-creator-5/)`
dsh 0.2.1-rc.2 install verified
`[![dsh.so install · dsh 0.2.1-rc.2](https://www.dsh.so/badge/install/dsh-skill-creator-5@0.2.1-rc.2.svg)](https://www.dsh.so/artifact/dsh-skill-creator-5/)`

Security Report

Two automated layers, neither a manual review: a static scan (per-plugin rule checks) and a model-driven dynamic security audit that covers only the subset static-flagged at warn or above and never executes plugin code.

dsh-skill-creator-5Automated review · daily

Static low, dynamic high — stricter wins → overall high.

static low→dynamic high→overall high
0 critical·0 high
highoverall · high
1static scan · vet primaryvet AST static analysis (npm artifact or git source)
Version binding
plugin versionv0.0.4dsh manifest
scanned version0.0.4npm2026-10-21
current version0.0.4same version
vet verdictclean · git

Vet is an AST static scan (npm artifact or git source).

show static findings (8)
  • warningchildProcess · code-execsrc/core/creators.mjs:101

    const clone = spawnSync('git', ['clone', '--depth', '1', '--branch', spec.branch, '--filter=blob:none', '--sparse', spec.repo, staging],

  • warningchildProcess · code-execsrc/core/creators.mjs:104

    const co = spawnSync('git', ['-C', staging, 'sparse-checkout', 'set', spec.path], { encoding: 'utf8', timeout: 120000 });

  • warningchildProcess · code-execsrc/core/creators.mjs:106

    const head = spawnSync('git', ['-C', staging, 'rev-parse', 'HEAD'], { encoding: 'utf8', timeout: 30000 });

  • warningfsWrite · filessrc/core/creators.mjs:149

    fs.writeFileSync(dest, file.data);

  • warningfsWrite · filessrc/core/creators.mjs:179

    fs.writeFileSync(stampPath(kind), `${JSON.stringify(stamp, null, 2)}\n`);

  • warningfsWrite · filessrc/core/plugin.mjs:36

    fs.writeFileSync(marker, `${JSON.stringify({ ...detail, at: new Date().toISOString(), pid: process.pid, cwd: process.cwd() }, null, 2)}\n`);

  • warningchildProcess · code-execsrc/core/runner.mjs:180

    result = spawnSync(bash, [script, '-s', skillDir], {

  • warningfsWrite · filessrc/core/runner.mjs:400

    fs.writeFileSync(tmp, `${JSON.stringify({ version: 1, runs: merged }, null, 1)}\n`);

2dynamic auditmodel reads redacted source context · never executes plugin code
highest gradehigh
merged intooverall high
audited at2026-10-19
findings0 critical·1 high1 medium / 1 low hidden
show 3 audit findings
  • highShipped config forces danger-full-access on unattended LLM agentscordis.patch.yml:49

    Every install applies this bundle patch, so runs created by POST /run get agents with sandbox=danger-full-access and no approval (runner.mjs spawnAgent sets the preset on the agent session). Those agents execute shell/CLI chosen by an LLM that is steered by a remotely fetched creator SKILL.md and by the request's free-text description; the 'readonly only' evaluation policy is prompt text, not enforcement. Normal use of the panel ('开始创建') therefore yields unsandboxed command execution with the DSH process's privileges and filesystem/network reach, and the plugin writes into <cwd>/skills (defaul

  • mediumCreator code is fetched from an unpinned branch head and executed with bashsrc/core/runner.mjs

    The validator script path is <cache>/<kind>/.../scripts/validate-skill.sh, produced by `git clone --depth 1 --branch spec.branch` (creators.mjs, branches 'master'/'main') with no commit pinning, signature or hash verification; the same unpinned creator tree is also handed to the creation agent with instructions to run its scripts. Any push to those upstream branches, or a compromised mirror/DNS path, becomes shell execution inside the DSH process (and, per the previous finding, full-access LLM execution) on the next evaluation. The no-git fallback apiWalk writes files at path.join(to, file.rel

  • lowRun history and mount marker written world-readable to a predictable temp pathsrc/core/creators.mjs

    runs.json (steps, spec/description text, target paths, cwd) and mounted.json (cwd, pid) are written under os.tmpdir()/dsh-skill-creator-cache with default mkdir/write modes (typically 0755/0644). On a multi-user host other local users can read pasted requirements and generated skill locations, and can plant a runs.json that the runner rehydrates and displays (React escapes the text, so impact is limited to disclosure/spoofed history).

Ships a default danger-full-access preset for unattended LLM agents and executes creator scripts fetched from unpinned branch heads; routes/validators were not harvestable.

Rule: the dynamic audit can only raise risk, never lower it (raise-only, never an exemption); a wrong raise is withdrawn by human review and the tier falls back to the static verdict. Overall verdict = the stricter of the two layers.

Disclaimer: the static scan and the dynamic security audit are both automated analysis, not a security guarantee. Always review what you install.

Activity

Last commit 2026-10-16 · activity: Active

• Repo created: 2026-10-15

• Stars: ★ 0 · Forks: 0

• Health: Active — committed within last 30 days

Source

GitHub: github.com/wasteoneslife/dsh-skill-creator

Was this page helpful?