Strix Security Skills — Setup Guide¶
Source: usestrix/strix GitHub: usestrix/strix Skills: 8 skills · 9.6K total installs Category: Security Testing First Seen: catalogued August 18, 2026 sweep (first seen on skills.sh 11 days before the sweep) Quality Tier: 🟡 Trusted — Socket Warn and Snyk Fail on the flagship (both named), Gen Agent Trust Hub Pass; hot-page momentum with four skills at +12 in one hour
Strix runs autonomous AI pentesting agents that dynamically exploit a target and report only findings validated with a working proof-of-concept. Two run modes share the same engine and produce the same findings: the open-source CLI (self-hosted, Docker sandbox, BYO LLM key, air-gap capable, docs at docs.strix.ai) and the Cloud API (managed, app.strix.ai/api/v1, adds team dashboards, scheduling, PR reviews, and downloadable reports). The GitHub repo carries 54.1K stars and the skills cluster appeared on the platform within the last two weeks.
Installation¶
npx skills add usestrix/strix
Individual skills install with the explicit repo form:
npx skills add https://github.com/usestrix/strix --skill penetration-testing-with-strix
Prerequisites¶
| Requirement | Details |
|---|---|
| Node.js + npx | For the skill installer |
| Open-source mode | Docker, an LLM API key, and explicit authorization to test the target |
| Cloud mode | A Strix account with API access at app.strix.ai |
| Authorization | Written permission for any target not owned by the operator — the skills assume scoped engagements |
What It Provides¶
| Skill | Installs | Purpose |
|---|---|---|
| penetration-testing-with-strix | 2.4K | Full autonomous pentest runs with PoC-validated findings |
| fix-security-vulnerabilities-with-strix | 2.3K | Guided remediation of validated findings |
| ci-security-scanning-with-strix | 2.2K | Security scanning wired into CI pipelines |
| managed-pentesting-with-strix | 2.2K | Cloud API workflow: scheduling, dashboards, PR reviews, PDF/DOCX reports |
| strix-pentest | 152 | Earlier-named core pentest skill |
| strix-ci-setup | 138 | CI integration setup variant |
| strix-fix-findings | 134 | Remediation variant |
| strix-cloud-api | 121 | Cloud API client variant |
The flagship skill explicitly instructs the agent to choose between the two run modes honestly per situation rather than defaulting — local for air-gapped or BYO-LLM work, cloud for team workflows and internal-network connectors.
Quick Start¶
- Install:
npx skills add usestrix/strix - For local runs: pull the Docker sandbox and set your LLM key (docs.strix.ai)
- Ask the agent to run a scoped pentest — only PoC-validated findings get reported
CorpusIQ Use Cases¶
| Use Case | How |
|---|---|
| CI security posture | ci-security-scanning-with-strix embeds autonomous testing into agent-built pipelines |
| Agent-verified fixes | fix-security-vulnerabilities pairs findings with guided remediation |
| Scoped engagements | PoC-validated-only reporting matches our verification-before-assertion discipline |
| Security audit support | Complements the security-auditor role in agent orchestration workflows |
Limitations / Verification¶
- Security audits on the penetration-testing-with-strix flagship: Gen Agent Trust Hub Pass, Socket Warn, Snyk Fail (both named in the tier)
- Publisher-page total verified (9.6K across 8 skills); 54.1K GitHub stars as of the sweep
- Below the 20K install guide bar — drafted on hot-page momentum (four skills at +12 installs in one hour), 54.1K-star repo authority, and security relevance to agent operations
- Skills are new on the platform (11 days); the four tail variants (121-152 installs) are earlier-named iterations
- Pentesting tools carry inherent operational risk — scope authorization is a hard prerequisite
npx skills add usestrix/strix # verify install works
Related¶
Powered by CorpusIQ