Pentest AI MCP Server

by 0xSteph• 1.8k•• Scanned•Claim this listing•Request removal
Security & Testingai penetration testingai pentestingai securityappsecbug bounty

Open-source AI pentester that proves every finding. Machine oracles re-run each exploit; verified bugs ship a proof capsule you can replay yourself.

Pentest AI is an open-source mcp server by 0xSteph, written mainly in Python, with 1,791 GitHub stars and the MIT license. Its code was last updated Oct 9, 2026.

Scan summary: scanned Nov 5, 2026 at commit 97bc34c. Seen able to run shell commands, connect to the internet. Scan policy

Scanned
Full scan at 97bc34c · Nov 5, 2026
Scanned Builds
Static code scanwarn
Secrets scanfail
Dependency audit143 known vulnerabilitiesfail
Manifestnonewarn
Buildpass
Boot & handshakefail
Network egressn/a
Reproducible (×3)n/a

Findings (80)

  • high
    exfiltration issue
    Details are available to the maintainer after claiming this listing.
  • high
    exfiltration issue
    Details are available to the maintainer after claiming this listing.
  • high
    exfiltration issue
    Details are available to the maintainer after claiming this listing.
  • high
    dropper issue
    Details are available to the maintainer after claiming this listing.
  • high
    exfiltration issue
    Details are available to the maintainer after claiming this listing.
  • high
    exfiltration issue
    Details are available to the maintainer after claiming this listing.
  • high
    exfiltration issue
    Details are available to the maintainer after claiming this listing.
  • high
    exfiltration issue
    Details are available to the maintainer after claiming this listing.
  • high
    exfiltration issue
    Details are available to the maintainer after claiming this listing.
  • high
    secrets issue
    Details are available to the maintainer after claiming this listing.
  • high
    secrets issue
    Details are available to the maintainer after claiming this listing.
  • high
    secrets issue
    Details are available to the maintainer after claiming this listing.
  • high
    secrets issue
    Details are available to the maintainer after claiming this listing.
  • high
    secrets issue
    Details are available to the maintainer after claiming this listing.
  • high
    secrets issue
    Details are available to the maintainer after claiming this listing.
  • high
    secrets issue
    Details are available to the maintainer after claiming this listing.
  • high
    secrets issue
    Details are available to the maintainer after claiming this listing.
  • high
    secrets issue
    Details are available to the maintainer after claiming this listing.
  • high
    dependencies issue
    Details are available to the maintainer after claiming this listing.
  • high
    dependencies issue
    Details are available to the maintainer after claiming this listing.
  • high
    dependencies issue
    Details are available to the maintainer after claiming this listing.
  • high
    dependencies issue
    Details are available to the maintainer after claiming this listing.
  • high
    dependencies issue
    Details are available to the maintainer after claiming this listing.
  • medium
    Contacts a hard-coded IP address instead of a domain name.
    agents/poc_validator/poc_agent.py:29
  • medium
    Contacts a hard-coded IP address instead of a domain name.
    engine/probes/web/ssrf_cloud_metadata.py:72
  • medium
    Contacts a hard-coded IP address instead of a domain name.
    engine/probes/web/ssrf_cloud_metadata.py:73
  • medium
    Contacts a hard-coded IP address instead of a domain name.
    engine/probes/web/ssrf_cloud_metadata.py:75
  • medium
    Contacts a hard-coded IP address instead of a domain name.
    engine/probes/web/ssrf_cloud_metadata.py:79
  • medium
    Turns off TLS certificate checking, allowing traffic interception.
    engine/scanners.py:151
  • medium
    Turns off TLS certificate checking, allowing traffic interception.
    engine/scanners.py:395
  • medium
    Turns off TLS certificate checking, allowing traffic interception.
    engine/scanners.py:454
  • medium
    Turns off TLS certificate checking, allowing traffic interception.
    engine/scanners.py:558
  • medium
    Turns off TLS certificate checking, allowing traffic interception.
    engine/scanners.py:587
  • medium
    Runs shell commands. Fine for some tools, but it means the agent can run anything your user can.
    scripts/diag_post_engagement_hang.py:117
  • medium
    aiohttp@3.13.5: 22 known vulnerabilities
    PYSEC-2026-2104, PYSEC-2026-2105, PYSEC-2026-2106, PYSEC-2026-2107, PYSEC-2026-2108 +17 more
    uv.lock
  • medium
    cryptography@48.0.0: 4 known vulnerabilities
    PYSEC-2026-3552, PYSEC-2026-3553, PYSEC-2026-3554, GHSA-537c-gmf6-5ccf
    uv.lock
  • medium
    dnspython@2.8.0: 1 known vulnerability
    GHSA-45c3-73f7-pv4m
    uv.lock
  • medium
    filelock@3.29.0: 9 known vulnerabilities
    GHSA-567q-5qvw-pf92, GHSA-69hj-f68j-2879, GHSA-8mx5-7m5m-r77g, GHSA-9vhv-jxww-xfm5, GHSA-cwqr-hh33-gh44 +4 more
    uv.lock
  • medium
    fsspec@2026.4.0: 1 known vulnerability
    PYSEC-2026-4217
    uv.lock
  • medium
    joserfc@1.6.5: 4 known vulnerabilities
    PYSEC-2026-2528, PYSEC-2026-2530, GHSA-9pg5-8c7x-5hfx, GHSA-jpxq-7h8p-2q76
    uv.lock
  • medium
    mcp@1.27.1: 7 known vulnerabilities
    PYSEC-2026-3481, PYSEC-2026-3482, PYSEC-2026-3483, GHSA-5h93-6whr-6q8j, GHSA-84m7-p3x7-pcfv +2 more
    uv.lock
  • medium
    protego@0.6.0: 1 known vulnerability
    PYSEC-2026-3906
    uv.lock
  • medium
    pyasn1@0.6.3: 7 known vulnerabilities
    PYSEC-2026-3455, PYSEC-2026-3456, PYSEC-2026-3457, GHSA-2qm8-x737-gmw5, GHSA-9m75-vc3g-4x39 +2 more
    uv.lock
  • medium
    python-multipart@0.0.28: 4 known vulnerabilities
    PYSEC-2026-3036, PYSEC-2026-3037, PYSEC-2026-3040, PYSEC-2026-3041
    uv.lock
  • medium
    soupsieve@2.8.3: 4 known vulnerabilities
    PYSEC-2026-3071, PYSEC-2026-3072, PYSEC-2026-4170, PYSEC-2026-4171
    uv.lock
  • medium
    starlette@1.0.0: 5 known vulnerabilities
    PYSEC-2026-161, PYSEC-2026-2280, PYSEC-2026-2281, PYSEC-2026-248, PYSEC-2026-249
    uv.lock
  • medium
    urllib3@2.7.0: 3 known vulnerabilities
    PYSEC-2026-4175, PYSEC-2026-4176, PYSEC-2026-4177
    uv.lock
  • medium
    virtualenv@21.3.1: 19 known vulnerabilities
    PYSEC-2026-4011, PYSEC-2026-4012, PYSEC-2026-4013, PYSEC-2026-4014, GHSA-394h-qwmc-48pf +14 more
    uv.lock
  • medium
    MCP server didn't complete the protocol handshake
    server exited early with code 2
  • low
    Possible example credential in docs/tests: jwt
    benchmarks/results/2026-05-15/run_summary.md:66
  • low
    Possible example credential in docs/tests: jwt
    benchmarks/results/2026-05-15/efd01c52_llm_coordinated_run.md:63
  • low
    Possible example credential in docs/tests: generic-api-key
    benchmarks/results/2026-05-12/juice-shop/triage.csv:92
  • low
    Possible example credential in docs/tests: jwt
    benchmarks/results/2026-05-12/juice-shop/ptai-sweep/unauthed-findings.json:983
  • low
    Possible example credential in docs/tests: generic-api-key
    benchmarks/results/2026-05-12/juice-shop/ptai-sweep/unauthed-findings.json:983
  • low
    Possible example credential in docs/tests: jwt
    2 matches in this file
    benchmarks/results/2026-05-15-juiceshop/c1ae665a_findings.json:1
  • low
    Possible example credential in docs/tests: generic-api-key
    benchmarks/results/2026-05-15-juiceshop/c1ae665a_findings.json:1
  • low
    Possible example credential in docs/tests: jwt
    2 matches in this file
    benchmarks/results/2026-05-15-juiceshop/2dd29513_findings.json:1
  • low
    Possible example credential in docs/tests: generic-api-key
    benchmarks/results/2026-05-15-juiceshop/2dd29513_findings.json:1
  • low
    Possible example credential in docs/tests: jwt
    benchmarks/results/2026-05-12/juice-shop/ptai-llm/unauthed-findings.json:983
  • low
    Possible example credential in docs/tests: generic-api-key
    benchmarks/results/2026-05-12/juice-shop/ptai-llm/unauthed-findings.json:983
  • low
    Possible example credential in docs/tests: jwt
    benchmarks/results/2026-05-12/juice-shop/ptai-sweep/authed-findings.json:983
  • low
    Possible example credential in docs/tests: generic-api-key
    benchmarks/results/2026-05-12/juice-shop/ptai-sweep/authed-findings.json:983
  • low
    Possible example credential in docs/tests: generic-api-key
    tests/test_oracle_oast.py:15
  • low
    Possible example credential in docs/tests: private-key
    tests/test_oob_client.py:219
  • low
    Possible example credential in docs/tests: generic-api-key
    tests/test_oracle_nosql.py:30
  • low
    Possible example credential in docs/tests: generic-api-key
    tests/test_oracle_sqli_login_bypass.py:32
  • low
    Possible example credential in docs/tests: jwt
    tests/test_report_draft.py:147
  • low
    Possible example credential in docs/tests: private-key
    tests/test_search_sqli_union_followup.py:100
  • low
    Possible example credential in docs/tests: generic-api-key
    tests/test_probe_nosql_fuzz.py:44
  • low
    Possible example credential in docs/tests: github-pat
    tests/test_oracle_leaked_credential.py:24
  • low
    Possible example credential in docs/tests: jwt
    tests/test_scanners.py:701
  • low
    fastmcp@3.2.4: 1 known vulnerability
    GHSA-7xgj-8392-5q4r
    uv.lock
  • low
    idna@3.13: 1 known vulnerability
    PYSEC-2026-215
    uv.lock
  • low
    multidict@6.7.1: 1 known vulnerability
    PYSEC-2026-4244
    uv.lock
  • low
    opentelemetry-api@1.41.1: 1 known vulnerability
    GHSA-xcq7-p3pf-p88h
    uv.lock
  • low
    pydantic-settings@2.14.1: 1 known vulnerability
    GHSA-4xgf-cpjx-pc3j
    uv.lock
  • low
    setuptools@82.0.1: 2 known vulnerabilities
    PYSEC-2026-3447, GHSA-xmj3-9gf7-h52w
    uv.lock
  • low
    werkzeug@3.1.8: 1 known vulnerability
    PYSEC-2026-4312
    uv.lock
  • low
    No manifest declaring network access and environment variables (agent.json)
  • info
    Ignored the repo's scanner suppression config (1 file)
    .gitleaks.toml

Network activity

During install
files.pythonhosted.orgpypi.org
While running
Not run — the code wasn't started in this scan

Automated result for the commit shown. See our scan policy. View the scan log

Maintainers can claim the listing to rescan, or dispute a result.

Pentest AI alternatives

View more →

Similar mcp servers in Security & Testing, ranked by GitHub stars. Compare licence, recent activity and scan status before you choose.

NameStarsLicenceLast updatedScan
Pentest AI (this page)
0xSteph
1.8kMITOct 9, 2026Scanned, no flagged issues
Hexstrike AI
0x4m4
13kMITAug 3, 2026Scanned, no flagged issues
Viper
FunnyWolf
5.3kNone foundMay 31, 2026Not scanned yet
Open Ontologies
fabio-rovai
2.7kMITOct 30, 2026Scanned, no flagged issues
Qu1cksc0pe
CYB3RMX
2.1kGPL-3.0Sep 18, 2026Scanned, no flagged issues
Cve MCP Server
mukul975
1.7kApache-2.0Oct 1, 2026Scanned, no flagged issues