Install verification: sandbox install ok, web smoke failed (L5) · tested on dsh 0.2.1-rc.2 · last checked Install verification details →
What it doesAI
VulnClaw pentest mode adds nmap, HTTP probing, directory/JS recon, traffic evidence, reports and PoCs as native DeepSeek Harness tools.
- Nmap scanning and HTTP probing as native Harness tools
- Directory and JavaScript recon for attack-surface discovery
- Traffic evidence capture plus pentest reports and PoC output
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:R1145141919810/dsh-vulnclaw-pentestInstall method: GitHub · sandbox install ok, web smoke failed (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 |
|---|---|---|
| 91bc56 | ✗ | ✗ |
91bc56github2 failed2026-10-19
https://github.com/R114514/dsh-vulnclaw-pentest
✗dsh 0.2.1-rc.2current2026-10-19L5 · failed
- • Node.js: v24.14.0
- • DSH: 0.2.1-rc.2
- • Plugin artifact: github · https://github.com/R1145141919810/dsh-vulnclaw-pentest (version unstated)
ID: dsh-vulnclaw-pentest@91bc56@dsh0.2.1-rc.2 · profile: l4-sandbox
Issued: 2026-10-19 · expires: 2026-10-26
View checks & evidence
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 · failed
- • Node.js: v24.14.0
- • DSH: 0.2.1-alpha.4
- • Plugin artifact: github · https://github.com/R1145141919810/dsh-vulnclaw-pentest (version unstated)
ID: dsh-vulnclaw-pentest@91bc56@dsh0.2.1-alpha.4 · profile: l4-sandbox
Issued: 2026-10-16 · expires: 2026-10-23
View checks & evidence
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-vulnclaw-pentest/)`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 medium — stricter wins → overall medium.
v1.1.0dsh manifest1.1.0git2026-10-211.1.0same versionclean · gitVet is an AST static scan (npm artifact or git source).
show static findings (6)
- warningchildProcess · code-exec
bridge.js:69const child = spawn(this.python, [this.script], { - warningnetIp · network
dev-regression.mjs:172url: "http://127.0.0.1:9527/login.php", - warningnetIp · network
dev-regression.mjs:194const attackErrors = accepts(SOURCE.attack_map, { findings: findingsCall.findings, target: "http://127.0.0.1:9527/" }); - warningnetIp · network
dev-verify-activation.mjs:184{ target: "http://127.0.0.1:9527/", goal: "activation regression" }, - warningnetIp · network
dev-verify-activation.mjs:190check("it reports the applied target", parsed.applied?.target === "http://127.0.0.1:9527/", text.slice(0, 300)); - warningchildProcess · code-exec
python-runtime.js:73const result = spawnSync(command, args, {
show 2 audit findings
- mediumshell_command/python_execute exposed as native tools with VulnClaw approval gate set to full_access
index.js:52Every VulnClaw tool is registered, including shell_command and python_execute. The bridge forces config.safety.permission_mode = full_access, disabling VulnClaw's own y/N approval. Content fetched from a target (pages, tool output) is untrusted and can prompt-inject the model into calling these execute-class tools, yielding arbitrary host command execution unless the Harness per-call approval reliably blocks/intercepts it.
- mediumFirst tool call installs unpinned vulnclaw[web] and requests from PyPI, then imports and runs them
python-runtime.js:35On first tool use the plugin runs `uv pip install` with no version pin or hash for the vulnclaw distribution and for requests, then imports vulnclaw inside the bridge process. A compromised upstream release or dependency confusion would execute arbitrary code in the managed venv on the host at runtime, not at a reviewable activation step.
Legitimate, well-structured pentest bridge; residual risk is host command execution reachable from prompt-injected content and unpinned PyPI installs at first use.
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: 2026-10-15
• Stars: ★ 0 · 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).