Install verification: runtime verified (L5) · tested on dsh 0.2.1-rc.2 · last checked Install verification details →
What it doesAI
Connects Unreal Engine 5.8+ built-in MCP server to DSH with one-click endpoint detection in settings, letting AI drive the UE editor.
- One-click detect and configure the UE MCP endpoint in DSH settings
- Replaces manual cordis.patch.yml YAML wiring
- Lets AI directly drive the Unreal Engine 5.8+ editor
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 github:yeee-eh/dsh-unreal-engineInstall method: GitHub · 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
○ The author declares no dsh version requirement — install is not blocked by version.
| plugin \ dsh | dsh 0.2.1-rc.2 current | dsh 0.2.1-alpha.4 |
|---|---|---|
| 08745c | ✓ | ✓ |
08745cgithub✓ 2 runtime-verified2026-10-19
https://github.com/yeee-eh/dsh-unreal-engine
✓dsh 0.2.1-rc.2current2026-10-19L5 · runtime verified
- • Node.js: v24.14.0
- • DSH: 0.2.1-rc.2
- • Plugin artifact: github · https://github.com/yeee-eh/dsh-unreal-engine (version unstated)
ID: dsh-unreal-engine@08745c@dsh0.2.1-rc.2 · profile: l4-sandbox
Issued: 2026-10-19 · expires: 2026-10-26
View checks & evidence
active in loader tree (151 entries)
L1–L3 are plugin-level static conclusions (baseline, identical across dsh versions): evidence is stored once per plugin.
✓dsh 0.2.1-alpha.42026-10-16L5 · runtime verified
- • Node.js: v24.14.0
- • DSH: 0.2.1-alpha.4
- • Plugin artifact: github · https://github.com/yeee-eh/dsh-unreal-engine (version unstated)
ID: dsh-unreal-engine@08745c@dsh0.2.1-alpha.4 · profile: l4-sandbox
Issued: 2026-10-16 · expires: 2026-10-23
View checks & evidence
active in loader tree (151 entries)
L1–L3 are plugin-level static conclusions (baseline, identical across dsh versions): evidence is stored once per plugin.
Author badges
Embed the official badges in your README to showcase security & install-test status:
`[](https://www.dsh.so/artifact/dsh-unreal-engine/)``[](https://www.dsh.so/artifact/dsh-unreal-engine/)`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.
Static low, dynamic low — stricter wins → overall low.
v0.1.0dsh manifest0.1.0git2026-10-210.1.0same versionclean · gitVet is an AST static scan (npm artifact or git source).
show static findings (9)
- criticaleval · code-exec
lib/js-expr-probe.mjs:7const evaluate = new Function('ctx', 'expr', 'return eval(expr)'); - criticalnewFunction · code-exec
lib/js-expr-probe.mjs:7const evaluate = new Function('ctx', 'expr', 'return eval(expr)'); - warningnetIp · network
lib/endpoint.mjs:63return { url: `http://127.0.0.1:${port}/mcp`, port, host: '127.0.0.1' }; - warningnetIp · network
lib/endpoint.mjs:71return { url: `http://127.0.0.1:${port}/mcp`, port, host: '127.0.0.1' }; - warningnetIp · network
lib/endpoint.mjs:102out.push(`http://127.0.0.1:${port}/mcp`); - warningnetIp · network
lib/js-expr-probe.mjs:6const expr = "process.env.DSH_UNREAL_MCP_URL || 'http://127.0.0.1:8000/mcp'"; - warningnetIp · network
lib/js-expr-probe.mjs:19check('未设 env 时回退到默认端点', evaluate({}, expr), 'http://127.0.0.1:8000/mcp'); - warningnetIp · network
lib/js-expr-probe.mjs:22process.env.DSH_UNREAL_MCP_URL = 'http://127.0.0.1:59999/mcp'; - warningnetIp · network
tools/ue5-dsh-doctor.mjs:19const MCP_URL = process.env.DSH_UNREAL_MCP_URL ?? 'http://127.0.0.1:8000/mcp';
show 3 audit findings
- loweval/new Function in a shipped dev probe script that the runtime never imports
lib/js-expr-probe.mjs:7This script reproduces the loader's !!js evaluation to prove process.env is visible in that scope. The evaluated expression is a hardcoded constant (line 6) and the file is only run by hand via `npm run probe`; no runtime module imports it, so no plugin request or untrusted input can reach it. It is a documentation/verification artifact, not a reachable code-execution path.
- lowBundle patch !!js expression lets an environment variable choose the MCP endpoint
cordis.patch.yml:68The loader evaluates this author-written static expression when applying the bundle; the result becomes the URL of the @deepseek-ai/dsh-mcp-client insert, i.e. whoever can set DSH_UNREAL_MCP_URL before DSH starts chooses the tool provider the agent talks to. There is no code injection: the env value is returned as a string, never spliced into code, and the expression itself is fixed in the package.
- lowDetect route opens outbound MCP connections to any host derived from the request body
lib/index.js:181POST /unreal-engine/detect accepts any http(s) URL or host:port text, makes the DSH host process connect to it, and lib/probe.mjs returns distinguishable ECONNREFUSED / TIMEOUT / ENOTFOUND / 404 outcomes — an outbound-request and coarse port-scan primitive. On success the URL is also persisted to unreal-engine.json, so the next start's MCP endpoint is attacker-chosen (tool hijack). Severity is limited because the route is registered alongside the authenticated settings API (the browser client relies on same-origin cookies) and nothing in this plugin weakens that check.
Static-scan criticals refuted: eval lives in a hand-run dev probe, !!js is a static env read; residual risk is low (detect route can probe arbitrary hosts if it were reachable without auth).
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 2026-10-15 · activity: Active
• Repo created: 2026-10-14
• Stars: ★ 2 · Forks: 0
• Health: Active — committed within last 30 days
Source
Author README
View on GitHub ↗The author's original README from the repository — fetched live, unedited (the English page loads the default README).
