CVE PoC Generator
Research a CVE by ID, generate a standalone Python proof-of-concept script, and produce a detailed vulnerability report.
Workflow
- NVD Lookup - Query NVD API v2.0 for the CVE ID (
python3 tools/nvd-lookup.py <CVE>). Extract the CVSS score/vector using the v4.0-primary ladder (v4.0 → v3.1 → v3.0 → v2.0; the tool's JSON_SUMMARY exposes cvss_version + cvss_vector), plus CWE IDs, CPE matches, advisory URLs, and patch links.
- Advisory Research - Deep-dive vendor advisories, GitHub security advisories, Exploit-DB, and published write-ups. Identify root cause, affected versions, and attack vector details.
- PoC Generation - Write a standalone Python script (
poc.py) that demonstrates the vulnerability safely. Follow the script standards in reference/poc-methodology.md.
- Report Generation - Write a comprehensive markdown report (
report.md) with metadata, root cause analysis, risk assessment, and remediation guidance.
NVD Data to Collect
| Field |
Source |
Usage |
| CVE ID |
NVD |
Primary identifier |
| CVSS Score + Vector (v4.0 primary; v3.1/v3.0/v2.0 fallback) |
NVD |
Risk scoring |
| CWE ID(s) |
NVD |
Vulnerability classification |
| CPE Matches |
NVD |
Affected products and versions |
| Advisory URLs |
NVD references |
Research sources |
| Patch Links |
NVD references / vendor |
Remediation guidance |
| Description |
NVD |
Vulnerability summary |
| Published / Modified dates |
NVD |
Timeline |
Output
{OUTPUT_DIR}/
artifacts/cve-pocs/CVE-XXXX-XXXXX/
poc.py # Standalone Python PoC script
reports/cve-pocs/CVE-XXXX-XXXXX/
report.md # Detailed vulnerability report
Invocation
/cve-poc-generator CVE-2024-XXXXX
The skill accepts a single CVE ID as argument. Multiple CVEs should be processed with separate invocations.
Rules
- Least harm - PoC scripts MUST demonstrate vulnerability without causing damage. Use detection/verification checks, not destructive payloads.
- Standalone scripts - PoC must run independently with only standard Python libraries plus
requests. No framework dependencies.
- Accurate scoring - Use the exact CVSS score and vector from NVD. Do not fabricate or estimate scores.
- Source attribution - Every claim in the report must cite its source (NVD, vendor advisory, CVE description).
- No emoji - Use text severity labels only (CRITICAL, HIGH, MEDIUM, LOW, INFORMATIONAL).
- Verified data only - Do not hallucinate CVE details. If NVD data is unavailable, state it explicitly.
- Safe defaults - PoC scripts must default to read-only, non-destructive operations. Any potentially harmful action requires explicit
--confirm flag.
1---
2name: cve-poc-generator
3description: CVE research, standalone PoC script and report generation. Given a CVE ID, researches NVD and advisories, generates a safe Python PoC, and writes a detailed vulnerability report.
4---
5
6# CVE PoC Generator
7
8Research a CVE by ID, generate a standalone Python proof-of-concept script, and produce a detailed vulnerability report.
9
10## Workflow
11
121. **NVD Lookup** - Query NVD API v2.0 for the CVE ID (`python3 tools/nvd-lookup.py <CVE>`). Extract the CVSS score/vector using the v4.0-primary ladder (v4.0 → v3.1 → v3.0 → v2.0; the tool's `JSON_SUMMARY` exposes `cvss_version` + `cvss_vector`), plus CWE IDs, CPE matches, advisory URLs, and patch links.
132. **Advisory Research** - Deep-dive vendor advisories, GitHub security advisories, Exploit-DB, and published write-ups. Identify root cause, affected versions, and attack vector details.
143. **PoC Generation** - Write a standalone Python script (`poc.py`) that demonstrates the vulnerability safely. Follow the script standards in `reference/poc-methodology.md`.
154. **Report Generation** - Write a comprehensive markdown report (`report.md`) with metadata, root cause analysis, risk assessment, and remediation guidance.
16
17## NVD Data to Collect
18
19| Field | Source | Usage |
20|-------|--------|-------|
21| CVE ID | NVD | Primary identifier |
22| CVSS Score + Vector (v4.0 primary; v3.1/v3.0/v2.0 fallback) | NVD | Risk scoring |
23| CWE ID(s) | NVD | Vulnerability classification |
24| CPE Matches | NVD | Affected products and versions |
25| Advisory URLs | NVD references | Research sources |
26| Patch Links | NVD references / vendor | Remediation guidance |
27| Description | NVD | Vulnerability summary |
28| Published / Modified dates | NVD | Timeline |
29
30## Output
31
32```
33{OUTPUT_DIR}/
34 artifacts/cve-pocs/CVE-XXXX-XXXXX/
35 poc.py # Standalone Python PoC script
36 reports/cve-pocs/CVE-XXXX-XXXXX/
37 report.md # Detailed vulnerability report
38```
39
40## Invocation
41
42```
43/cve-poc-generator CVE-2024-XXXXX
44```
45
46The skill accepts a single CVE ID as argument. Multiple CVEs should be processed with separate invocations.
47
48## Rules
49
501. **Least harm** - PoC scripts MUST demonstrate vulnerability without causing damage. Use detection/verification checks, not destructive payloads.
512. **Standalone scripts** - PoC must run independently with only standard Python libraries plus `requests`. No framework dependencies.
523. **Accurate scoring** - Use the exact CVSS score and vector from NVD. Do not fabricate or estimate scores.
534. **Source attribution** - Every claim in the report must cite its source (NVD, vendor advisory, CVE description).
545. **No emoji** - Use text severity labels only (CRITICAL, HIGH, MEDIUM, LOW, INFORMATIONAL).
556. **Verified data only** - Do not hallucinate CVE details. If NVD data is unavailable, state it explicitly.
567. **Safe defaults** - PoC scripts must default to read-only, non-destructive operations. Any potentially harmful action requires explicit `--confirm` flag.
transilienceai