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 soc-metricsgit 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/soc-metrics)<a href="https://agentmods.dev/skills/gensecaihq/wazuh-autopilot/soc-metrics"><img src="https://agentmods.dev/badge/skills/gensecaihq/wazuh-autopilot/soc-metrics/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/soc-metrics"><img src="https://agentmods.dev/badge/skills/gensecaihq/wazuh-autopilot/soc-metrics.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 | $0.00056 | $0.00859 |
| Opus 5.5 | $0.00022 | $0.00344 |
| Sonnet 5 | $0.00011 | $0.00172 |
| Haiku 4.5 | $0.00006 | $0.00086 |
Grade A, and why
soc-metrics 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 — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SOC Metrics
Metrics should drive improvement, not vanity. Define precisely, compute consistently, and always show the trend and the sample size.
Time metrics (per incident, then median and p90 over the period)
| Metric | Definition | Start → End |
|---|---|---|
| MTTD — mean time to detect | how long the attacker was active before detection | first malicious activity → first alert |
| MTTA — mean time to acknowledge | time until someone (agent or human) started work | case created → first triage finding |
| MTTC — mean time to contain | time to stop the harm | case created → containment verified |
| MTTR — mean time to respond/resolve | time to close out | case created → resolved/closed |
Report median and p90 alongside the mean — a few long incidents skew means. MTTD requires an incident timeline; mark it "n/a" where the start of malicious activity is unknown.
Volume and quality
| Metric | Formula | Healthy signal |
|---|---|---|
| Alert-to-incident ratio | alerts ingested ÷ cases opened | falling over time with stable coverage |
| False-positive rate | cases closed false_positive ÷ cases closed |
trending down; > 50% means tuning needed |
| Escalation rate | cases reaching investigation ÷ cases opened | stable, explainable |
| Automation rate | actions auto-approved or auto-executed ÷ all executed actions | increases only with low rollback rate |
| Rollback rate | actions rolled back ÷ executed | < 5%; spikes mean over-aggressive autonomy |
| Verification rate | executed actions verified ÷ executed | → 100% |
| Approval latency | action proposed → approved (median) | within policy expiry |
| Agent coverage | active Wazuh agents ÷ expected assets | → 100% |
| Telemetry loss | alerts for rule 203 (event queue full) or rule 204 (queue flooded), and agents disconnected for > 1h (rule 504) | zero; any non-zero means detection blind spots. Get detail from platform-engineer |
| MTTD | first attacker event time (from the log, not the alert timestamp) → case opened |
falling; ingestion delay shows up as the gap between event time and alert time |
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 · 72 lines · 56 tokens per session scan A 6c9f5f05bdd4
soc-metrics is a skill published in the GitHub repository gensecaihq/Wazuh-Autopilot (57 stars, last pushed 2d ago), licensed MIT. It adds 56 tokens to every session and 859 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.