techstack-identification — Claude Code skill for stack discovery
techstack-identification is a Claude Code skill that finds a company’s technology stack from public signals. It delegates research to specialist sub-skills and combines their findings in a report.
MIT 215 installs ★ 632 updated 2026-07-29
npx skills add transilienceai/communitytools --skill techstack-identification -a claude-codeWhat techstack-identification does
techstack-identification investigates a company’s stack using information available to the public. It starts with infrastructure, then examines frontend, backend, security, and OSINT signals; a correlation sub-skill checks how the findings fit together and assigns confidence levels.
The final report comes in JSON and Markdown, with discovered assets, technologies, versions, and supporting evidence. Other workflows cover CDN/WAF discovery, API mapping, CMS fingerprinting, and SaaS exposure. Research stays passive: no active scans, credentials, DNS zone transfers, or brute force. External requests must be logged, and the listed sources have rate limits.
What it helps with
- identify a company’s web technologies from public signals
- inventory a company’s domains, subdomains, and IP addresses
- compare frontend and backend indicators
- score confidence in identified technologies
- find public evidence of SaaS and infrastructure exposure
Example prompts for the agent
- Identify the stack for
example.comand return a report. - Build an inventory of the company’s infrastructure from public sources.
- Find CMS and API signals for the company, then correlate the evidence.
How to install techstack-identification
1Skills CLI (Vercel)
npx skills add transilienceai/communitytools --skill techstack-identification -a claude-code2Manually: put the files in place
~/.claude/skills/techstack-identification/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/techstack-identification/
1Skills CLI (Vercel)
npx skills add transilienceai/communitytools --skill techstack-identification -a codex2Manually: put the files in place
.agents/skills/techstack-identification/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add transilienceai/communitytools --skill techstack-identification -a cursor2Manually: put the files in place
.cursor/skills/techstack-identification/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install transilienceai/communitytools skills/techstack-identification2Manually: put the files in place
.github/skills/techstack-identification/SKILL.md1Skills CLI (Vercel)
npx skills add transilienceai/communitytools --skill techstack-identification -a gemini-cli2Platform CLI
gemini skills install https://github.com/transilienceai/communitytools --path skills/techstack-identificationManually: put the files in place
.devin/skills/techstack-identification/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add transilienceai/communitytools --skill techstack-identification -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about techstack-identification
How do I install techstack-identification in Claude Code?
Run npx skills add transilienceai/communitytools --skill techstack-identification -a claude-code. The skill file is at ~/.claude/skills/techstack-identification/SKILL.md.
What does techstack-identification do?
It investigates a company’s technology stack through public sources, delegates research to six specialist sub-skills, and compiles the findings into JSON and Markdown reports.
How does techstack-identification handle reconnaissance?
It uses passive methods and public sources only. Active scanning, credentialed access, zone transfers, and brute force are excluded; external requests are logged.
Source text
SKILL.md · MIT
---
name: techstack-identification
description: OSINT-based technology stack identification. Routes to 6 domain sub-skills (frontend, backend, infra, security, osint, correlation) to discover a target's stack from publicly available signals.
---
# Tech Stack Identification
Passive OSINT reconnaissance to identify a target's technology stack. No credentials, no active scanning — only publicly available signals.
## Quick Start
```
1. Provide company name (+ optional domain hint)
2. Run infra first (asset inventory) → frontend / backend / security / osint in parallel
3. Pass all signals into correlation → final report (JSON + Markdown)
```
## Domain Sub-Skills
| Sub-skill | Identifies | Read |
|-----------|-----------|------|
| **frontend** | JS frameworks, meta-frameworks, CSS libraries, build tools, CMS via DOM/HTML/JS | [frontend/SKILL.md](frontend/SKILL.md) |
| **backend** | Web servers, runtimes, languages, frameworks, DB, APIs, CMS | [backend/SKILL.md](backend/SKILL.md) |
| **infra** | Cloud, CDN/WAF, DNS, TLS/CT, DevOps, asset discovery (domains/subdomains/IPs) | [infra/SKILL.md](infra/SKILL.md) |
| **security** | Security headers, CSP, email auth, security.txt, third-party SaaS | [security/SKILL.md](security/SKILL.md) |
| **osint** | Public repos (GitHub/GitLab), job postings/ATS, Wayback Machine | [osint/SKILL.md](osint/SKILL.md) |
| **correlation** | Cross-validation, confidence scoring, conflict resolution | [correlation/SKILL.md](correlation/SKILL.md) |
## Routing by Objective
| Objective | Mount |
|-----------|-------|
| Full stack discovery | infra → (frontend, backend, security, osint) → correlation |
| CDN/WAF identification only | infra |
| API surface mapping | backend |
| Supply-chain / SaaS exposure | security + osint |
| CVE matching by version | backend + frontend (then correlation) |
| Migration / historical context | osint (web archive) + correlation |
| CMS fingerprint | frontend (HTML generators) + backend (CMS paths/cookies) |
| Asset inventory only | infra (domain discovery, subdomain enum, IP attribution, CT) |
## Confidence Levels
- **High**: 3+ independent sources OR explicit identifier (header/meta/global) + supporting evidence + version known
- **Medium**: Single strong source OR multiple indirect signals (URL patterns, cookies, DOM attrs, job postings)
- **Low**: Speculative — single weak signal, conflicting data, or archive-only evidence
Computed in `correlation/`. Target distribution: 50-70% High, 20-35% Medium, <15% Low.
## Final Report Schema
```json
{ "report_id": "uuid", "company": "string", "primary_domain": "string",
"discovered_assets": {"domains", "subdomains", "ip_addresses", "certificates", "api_portals"},
"technologies": {
"frontend": [{"name", "version?", "confidence", "evidence": []}],
"backend": [...], "infrastructure": [...], "security": [...],
"devops": [...], "third_party": [...] },
"confidence_summary": {"high_confidence", "medium_confidence", "low_confidence", "overall_score"} }
```
## Rate Limits
crt.sh 10/min · GitHub (unauth) 60/h · HTTP 30/min/domain · DNS 30/min · Wayback CDX 15/min · WHOIS 5/min.
## Ethics
Passive only. No active scanning, credentialed access, zone transfers, or brute force. Public sources only. Log every external request for audit.
© transilienceai · MIT · transilienceai/communitytools