Install verification: runtime verified (L5) · tested on dsh 0.2.1-rc.2 · last checked Install verification details →
What it doesAI
在 DSH(DeepSeek Harness)设置页里创建/编辑 agent 预设的可视化工作台:分段提示词 + 能力开关 + 内置模板。A visual agent-preset workbench for DSH.
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 dsh-preset-workbench@1.1.0security scan passed · npm package v1.1.0
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
○ 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 |
|---|---|---|
| 1.0.2 | · | · |
| v1.0.0 | ✓ | ✓ |
1.0.2npm2026-10-21
dsh-preset-workbench
No runtime install test for this version (static scan only) — compatibility unknown, never assumed.
v1.0.0github✓ 2 runtime-verified2026-10-19
https://github.com/DDDMUC/dsh-preset-workbench
✓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/DDDMUC/dsh-preset-workbench#v1.0.0
ID: dsh-preset-workbench@v1.0.0@dsh0.2.1-rc.2 · profile: l4-sandbox
Issued: 2026-10-19 · expires: 2026-10-26
View checks & evidence
↳ carried from earlier record dsh-preset-workbench@1.0.2@dsh0.2.1-rc.2 (l3-fullscan)
↳ carried from earlier record dsh-preset-workbench@1.0.2@dsh0.2.1-rc.2 (l3-fullscan)
↳ carried from earlier record dsh-preset-workbench@1.0.2@dsh0.2.1-rc.2 (l3-fullscan)
active in loader tree (150 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/DDDMUC/dsh-preset-workbench#v1.0.0
ID: dsh-preset-workbench@v1.0.0@dsh0.2.1-alpha.4 · profile: l4-sandbox
Issued: 2026-10-16 · expires: 2026-10-23
View checks & evidence
↳ carried from earlier record dsh-preset-workbench@1.0.2@dsh0.2.1-rc.2 (l3-fullscan)
↳ carried from earlier record dsh-preset-workbench@1.0.2@dsh0.2.1-rc.2 (l3-fullscan)
↳ carried from earlier record dsh-preset-workbench@1.0.2@dsh0.2.1-rc.2 (l3-fullscan)
active in loader tree (150 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-preset-workbench/)``[](https://www.dsh.so/artifact/dsh-preset-workbench/)`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.
v1.1.0dsh manifest1.1.0npm2026-10-211.1.0same versionclean · gitVet is an AST static scan (npm artifact or git source).
show 3 audit findings
- lowYAML parser forwards !!js expressions to the host evaluator (no plugin-side eval)
lib/yaml-lite.js:59A preset composition file containing `!!js <expression>` is parsed into a `{ __jsExpr }` row that the host's preset registry evaluates by itself, so a preset pack obtained from an untrusted source can carry JS into that host evaluator. This plugin never calls eval/Function and `!!js` is the host's own documented YAML dialect (the header comment says the host recognises and evaluates it), so the plugin adds no execution primitive of its own - the residual risk is inherited from the host feature.
- lowOpt-in 'system-prompt' mode lifts repo AGENTS.md text into the system prompt and condenses harness instruction messages
templates/liangshen/minimal-prompt.mjsIn `instructionSource: 'system-prompt'` mode, AGENTS.md / CLAUDE.md files discovered from $DSH_HOME and from the project root down to the cwd (1 MiB cap per file) become a system-prompt section, while the harness's equivalent user-role messages are replaced by a short marker (filterInstructionMessages: `if (message?.source?.baseline !== true) continue`). File text therefore reaches the model with system-prompt authority rather than user-message authority - relevant when a cloned third-party repo ships an AGENTS.md. Reads are confined to the project root and $DSH_HOME, and the mode is opt-in (d
- lowPreset write/delete and OS folder-open live in lib/index.js, which was not part of the audited source
package.json:6The client shown here drives state-changing operations (`save`, `create`, `delete`, `reveal`, `setBinding`) over POST /api/preset-workbench with client-supplied `id` and section `file` values, but the handler that turns them into filesystem writes/deletes and an OS 'open folder' action is absent from the harvested files, so preset-directory confinement, path validation and any origin/CSRF check on that loopback bridge are unverified. No evidence of a flaw - only of an unaudited surface.
No reachable code execution, exfiltration or shell use in the workbench UI, YAML parser or preset templates; the scanner's critical hit is refuted and only low hardening gaps remain.
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 2026-10-21 · activity: Active
• Repo created: 2026-10-15
• Stars: ★ 1 · 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).
