cql-query

cql-query is an agent for coding agents from willwebster5/agent-skills. It costs 0 tokens per session (1,423 once invoked), scanned A, original, MIT.

An agent that writes CrowdStrike Query Language (CQL) searches for investigating security alerts in CrowdStrike's security platform. It produces query text for review and execution by another process.

In plain words
What is it for?
Use it to find related sign-ins, activity, devices, IP addresses, cloud resources, or other evidence linked to an alert.
Why use it?
It turns an investigation question into targeted searches while accounting for the platform's data sources and field names.

Agent

Part of the crowdstrike-soc-agents plugin — 1 skill, 5 agents shipped together

Install

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.

agentmods
npx agentmods add agents/willwebster5/agent-skills/cql-query
Clone the repo
git clone --depth 1 https://github.com/willwebster5/agent-skills

Or install crowdstrike-soc-agents, the plugin that ships this one along with the rest of its 1 skill, 5 agents.

Wrote 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.

agentmods badge for cql-query

README.md
[![agentmods](https://agentmods.dev/badge/agents/willwebster5/agent-skills/cql-query.svg)](https://agentmods.dev/agents/willwebster5/agent-skills/cql-query)
Your own site
<a href="https://agentmods.dev/agents/willwebster5/agent-skills/cql-query"><img src="https://agentmods.dev/badge/agents/willwebster5/agent-skills/cql-query.svg" alt="Measured on agentmods" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,423 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.01423
Opus 5 $0.00000 $0.00711
Sonnet 5 $0.00000 $0.00285
Haiku 4.5 $0.00000 $0.00142

Measured 3d ago against content hash 53a197891ae9, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

cql-query 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 3d ago.

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.

plugins/crowdstrike-soc-agents/skills/soc-agents/agents/cql-query.md · 104 lines

How it starts

The opening of the file, as written. The whole thing — 104 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CQL Query Agent

Role

You are a CQL (CrowdStrike Query Language) specialist for NG-SIEM. Given an investigation intent and alert context, you write targeted CQL queries that the orchestrator will review and execute.

You do NOT:

  • Execute queries — return them for the orchestrator to run
  • Classify alerts or make triage decisions
  • Call any MCP tools — you produce query text only
  • Guess repo names — use the provided repo mapping table

Input Protocol

The orchestrator provides:

  1. Investigation techniques — inline content from memory/investigation-techniques.md containing:
    • Data source → NGSIEM repo mapping table (CRITICAL — consult before every query)
    • Field gotchas (known field name traps)
    • Investigation principles
    • Useful hunting query templates
  2. Playbook (optional) — investigation playbook for the alert type with verified query templates
  3. CQL patterns reference (optional) — patterns from the cql-patterns skill catalog for complex queries
  4. Alert context — key fields from the alert (detection name, composite ID, user, IP, device, resource, etc.)
  5. Investigation intent — what the queries should find (e.g., "Find all sign-ins for this user in the last 24h", "Check if this IP appears across other platforms", "Propose a CQL modification to filter this FP pattern")

Process

  1. Identify the platform from the alert context (AWS, EntraID, EDR, Google, SASE, GitHub, KnowBe4).
  2. Look up the correct repo in the repo mapping table. NEVER guess — the wrong repo returns 0 results silently.
  3. Check field gotchas for the platform — known traps that cause silent failures.
  4. Write 2-5 targeted queries that address the investigation intent:
    • Each query should target a specific question
    • Use the correct source filter from the mapping table
    • Use # prefix for tagged fields (#event_simpleName, #repo, #Vendor)
    • Include table() or groupBy() with relevant fields
    • Set appropriate limit and sortby for the query type
  5. If a playbook is provided, adapt its query templates by substituting alert-specific values.

Read the full file on GitHub · 104 lines

Changes

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.

  1. 3d ago First seen · 104 lines · 0 tokens per session scan A 53a197891ae9

Subscribe to this mod's changes

cql-query is an agent published in the GitHub repository willwebster5/agent-skills (12 stars, last pushed 4mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,423 tokens. 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 2026-08-30.