secops-hunt

A guide for proactively searching security data for signs of hidden attacks.

In plain words
What is it for?
Use it to hunt for suspicious files, domains, IP addresses, URLs, campaigns, threat actors, and related attack patterns in security logs.
Why use it?
It helps security teams investigate possible threats by checking activity against indicators and attacker techniques instead of waiting for an alert.

Skill for Claude CodeCodex

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 skills/google/mcp-security/hunt
Any agent
npx skills add google/mcp-security --skill hunt
Clone the repo
git clone --depth 1 https://github.com/google/mcp-security

Made for: Claude Code, Codex.

Per session 34 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,142 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 $0.00034 $0.01142
Opus 5 $0.00017 $0.00571
Sonnet 5 $0.00007 $0.00228
Haiku 4.5 $0.00003 $0.00114

Measured 2d ago against content hash 0e084fb450cf, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

secops-hunt 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 2d 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.

Origin

Copies of this mod

1 near-identical copy found in the catalogue:

extensions/google-secops/skills/hunt/SKILL.md · 88 lines

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.

Threat Hunter

You are an expert Threat Hunter. Your goal is to proactively identify undetected threats in the environment.

Tool Selection & Availability

CRITICAL: Before executing any step, determine which tools are available in the current environment.

  1. Check Availability: Look for Remote tools (e.g., udm_search, get_ioc_match) first. If unavailable, use Local tools (e.g., search_security_events, get_ioc_matches).
  2. Reference Mapping: Use extensions/google-secops/TOOL_MAPPING.md to find the correct tool for each capability.
  3. Adapt Workflow: If using Remote tools for Natural Language Search, perform translate_udm_query then udm_search. If using Local tools, use search_security_events directly.

Procedures

Select the most appropriate procedure from the options below.

Proactive Threat Hunting based on GTI Campaign/Actor

Objective: Given a GTI Campaign or Threat Actor Collection ID (${GTI_COLLECTION_ID}), proactively search the local environment (SIEM) for related IOCs and TTPs.

Workflow:

  1. Analyst Input: Hunt for Campaign/Actor: ${GTI_COLLECTION_ID}
  2. IOC Gathering: Ask user for list of IOCs (files, domains, ips, urls) associated with the campaign/actor.
  3. Initial Scan:
    • Action: Check for recent hits against these indicators.
    • Remote: get_ioc_match.
    • Local: get_ioc_matches.
  4. Phase 1 Lookup (Iterative SIEM Search):
    • For each prioritized IOC, construct and execute the appropriate UDM query:
    • IP: principal.ip = "IOC" OR target.ip = "IOC" OR network.ip = "IOC"
    • Domain: principal.hostname = "IOC" OR target.hostname = "IOC" OR network.dns.questions.name = "IOC"
    • Hash: target.file.sha256 = "IOC" OR target.file.md5 = "IOC" OR target.file.sha1 = "IOC"
    • URL: target.url = "IOC"
    • Tool: udm_search (Remote/Local).
  5. Phase 2 Deep Investigation (Confirmed IOCs):
    • Action: Search SIEM events for confirmed IOCs to understand context (e.g. process execution, network connections).
    • Action: Check for related cases (list_cases).
  6. Synthesis: Synthesize all findings.
  7. Output: Ask user to Create Case, Update Case, or Generate Report.
    • If Report: Generate a markdown report file using write_file.
    • If Case: Post a comment to SOAR.

Read the full file on GitHub · 88 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. 2d ago First seen · 88 lines · 34 tokens per session scan A 0e084fb450cf

Subscribe to this mod's changes

secops-hunt is a skill published in the GitHub repository google/mcp-security (522 stars, last pushed 2d ago), licensed Apache-2.0. It adds 34 tokens to every session and 1,142 once invoked, about $0.0002 per session on Opus 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 2026-08-30.