Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add gensecaihq/Wazuh-Autopilot --skill severity-scoringgit clone --depth 1 https://github.com/gensecaihq/Wazuh-AutopilotWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/gensecaihq/wazuh-autopilot/severity-scoring)<a href="https://agentmods.dev/skills/gensecaihq/wazuh-autopilot/severity-scoring"><img src="https://agentmods.dev/badge/skills/gensecaihq/wazuh-autopilot/severity-scoring/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/gensecaihq/wazuh-autopilot/severity-scoring"><img src="https://agentmods.dev/badge/skills/gensecaihq/wazuh-autopilot/severity-scoring.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00047 | $0.01082 |
| Opus 5.5 | $0.00019 | $0.00433 |
| Sonnet 5 | $0.00009 | $0.00216 |
| Haiku 4.5 | $0.00005 | $0.00108 |
Grade A, and why
severity-scoring scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Severity Scoring
Severity must be explainable. Use this formula, then sanity-check it.
Inputs (each scored 0–10)
| Factor | Weight | How to score |
|---|---|---|
| Rule level (L) | 0.30 | min(10, rule.level * 10 / 15) using the highest level in the case (see note below) |
| Asset criticality (A) | 0.25 | 10 crown-jewel / domain controller / prod DB; 7 prod server; 4 internal workstation; 2 lab/test |
| Tactic weight (T) | 0.20 | see table below, highest tactic present |
| Blast radius (B) | 0.15 | 1 host = 2; 2–5 hosts = 5; >5 hosts or domain-wide identity = 9; org-wide = 10 |
| Evidence strength (E) | 0.10 | 10 confirmed success (e.g. login after brute force, file dropped); 5 attempt; 2 anomaly only |
score = 0.30L + 0.25A + 0.20T + 0.15B + 0.10E
About Wazuh levels
Wazuh levels describe the rule, not your environment. Some are misleading in context:
- High-level rules can be environmental or operational. For example, rule 204 (agent event queue flooded, level 12) is a pipeline problem, and rule 1003 (oversized syslog message, level 13) is often a misbehaving device. Keep L from the level, but score E low unless there is attacker evidence.
- Low-level rules can be the key evidence. For example, rule 5715 (sshd authentication success, level 3) right after rule 5712 (brute force, level 10) from the same source is a likely compromise. Score E = 10 and use the highest level in the case, not the success event's.
- Vulnerability-detector levels follow CVE severity (rules 23503, 23504, 23505, 23506 =
Low, Medium, High, Critical). That is exposure, not activity: route to
vulnerability-prioritizationinstead of scoring it as an intrusion.
Tactic weights (MITRE ATT&CK)
| Tactic | Weight |
|---|---|
| Reconnaissance, Resource Development | 2 |
| Initial Access (attempt), Discovery | 4 |
| Execution, Persistence, Defense Evasion | 6 |
| Credential Access, Privilege Escalation | 7 |
| Lateral Movement, Command and Control | 8 |
| Collection, Exfiltration | 9 |
| Impact | 10 |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 88 lines · 47 tokens per session scan A 5ae1b5860c23
severity-scoring is a skill published in the GitHub repository gensecaihq/Wazuh-Autopilot (57 stars, last pushed 2d ago), licensed MIT. It adds 47 tokens to every session and 1,082 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-25.
Other skills, from other repositories
agent-crdt-synchronizer
Agent skill for crdt-synchronizer - invoke with $agent-crdt-synchronizer.
agent-topology-optimizer
Agent skill for topology-optimizer - invoke with $agent-topology-optimizer.
agent-benchmark-suite
Agent skill for benchmark-suite - invoke with $agent-benchmark-suite.
agent-performance-monitor
Agent skill for performance-monitor - invoke with $agent-performance-monitor.
agent-resource-allocator
Agent skill for resource-allocator - invoke with $agent-resource-allocator.
agentic-jujutsu
Quantum-resistant, self-learning version control for AI agents with ReasoningBank intelligence and multi-agent coordination.