procmon

A Windows inspection tool that records what programs and their child processes do with files, the registry, networks, and processes. It saves the recording in a Procmon-compatible PML file, which can then be queried as JSON.

In plain words
What is it for?
Use it to investigate startup behavior, file writes, registry changes, network connections, process trees, and call stacks. It also helps filter and inspect events in PML recordings.
Why use it?
It helps replace guesswork about a program's behavior with a record of the actions it actually performed. Existing recordings can be analyzed without running a new capture.

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/progmboy/openprocmon/procmon
Any agent
npx skills add progmboy/openprocmon --skill procmon
Clone the repo
git clone --depth 1 https://github.com/progmboy/openprocmon

Made for: Claude Code, Codex.

Per session 80 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,968 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 $0.00080 $0.02968
Opus 5 $0.00040 $0.01484
Sonnet 5 $0.00016 $0.00594
Haiku 4.5 $0.00008 $0.00297

Measured 3d ago against content hash bbaa012c4d97, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

procmon 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.

skills/procmon/SKILL.md · 230 lines

How it starts

The opening of the file, as written. The whole thing — 230 lines — stays where its author put it; the contents beside it link to each section on GitHub.

OpenProcMon (procmon-cli)

OpenProcMon is a capture-then-analyze tool. A capture records process/file/ registry/network events to a Procmon-compatible .PML file; every analysis command reads a .PML and prints JSON. The same procmon-cli binary also serves these as MCP tools (procmon-cli mcp).

Prerequisites

  • 64-bit Windows. procmon-cli on PATH.
  • Live capture needs Administrator (it loads a kernel driver). Analyzing an existing .PML needs no elevation. Check with procmon-cli driver-status.

Workflow

  1. Capture the target's behavior to a .PML (or open an existing one).
  2. Query the .PML with filters to extract what you need.
  3. Drill in to a specific event for its detail + call stack.

1. Capture

Capture a program and all its children for 10s, launching it first:

procmon-cli capture --name notepad.exe --launch "notepad.exe" --duration 10
  • --name X (repeatable) targets a process by image name — matches present and future processes of that name. --pid N targets a specific instance.
  • Child processes are followed by default; --no-children to disable.
  • --launch "<cmd>" starts the program first so startup is captured.
  • --monitor process,file,registry,network selects sources (default: all).
  • --duration <secs> and --max-mb <N> bound the capture; --out <path> sets the .PML (default: a temp file). --filter "<expr>" narrows what's recorded (lossy — prefer narrowing at analysis time; same syntax as query below).
  • Empty --name/--pid captures the whole system. For injection-suspected malware (the payload may run inside an existing process, not a child), capture system-wide and filter afterward.

capture prints the pml_path, a summary, and a sample of events. Re-analyze that .PML with the commands below.

2. Query — the universal tool

procmon-cli query --pml <file> [--filter "<expr>"] [--group-by <cols>] [--metric <col>] [--limit N]

  • --filter "<expr>" — a single filter expression (see the Filter syntax section below). One flag expresses AND, OR, NOT and parens.
  • --group-by <Column[,Column...]> — return distinct values + counts (de-duplicated) instead of raw events. Comma-separate for multi-column (e.g. ProcessName,Path). Use it to avoid flooding (distinct files, not 5000 WriteFile rows).
  • --metric <numeric column/field> — roll up sum/avg/min/max + first/last time per group (e.g. --metric NetBytes for bytes per endpoint). Only with --group-by.
  • Noise (NTFS metadata, monitoring tools, IRP/FastIO bookkeeping, the tool itself) is excluded by default; add --no-noise to see everything.

Read the full file on GitHub · 230 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 · 230 lines · 80 tokens per session scan A bbaa012c4d97

Subscribe to this mod's changes

procmon is a skill published in the GitHub repository progmboy/openprocmon (332 stars, last pushed 6d ago), licensed MIT. It adds 80 tokens to every session and 2,968 once invoked, about $0.0004 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.

Related

Other skills, from other repositories

Driver Analysis

Windows kernel driver analysis — DriverEntry, dispatch table, IOCTL handlers, vulnerability audit.

buzzer-re/Rikugan · 20 tokens

wxjava-module-selector

根据微信公众号、小程序、微信支付、企业微信、开放平台、视频号或微信小店、腾讯企点和微信智能对话等业务场景,为用户选择合适的 WxJava Maven 模块、BOM 和示例入口。适用于用户询问“该用哪个模块”、依赖坐标、产品边界或单/多账号 Starter 选择时。.

binarywang/WxJava · 86 tokens

wxjava-api-contributor

按 WxJava 的 Maven 多模块、Java 8、公共 API 兼容性和 TestNG 约定,为微信官方接口新增或维护 SDK 支持。适用于新增 Service API、请求响应 Bean、序列化、HTTP 实现、Starter 配置或回归测试时。.

binarywang/WxJava · 68 tokens

azure-resource-manager-postgresql-dotnet

Azure PostgreSQL Flexible Server SDK for .NET. Database management for PostgreSQL Flexible Server deployments. Use for creating servers, databases, firewall rules, configurations, backups, and high availability. Triggers: "PostgreSQL", "PostgreSqlFlexibleServer", "PostgreSQL Flexible Server", "Azure Database for…

microsoft/skills · 97 tokens

azure-mgmt-apicenter-dotnet

Azure API Center SDK for .NET. Centralized API inventory management with governance, versioning, and discovery. Use for creating API services, workspaces, APIs, versions, definitions, environments, deployments, and metadata schemas. Triggers: "API Center", "ApiCenterService", "ApiCenterWorkspace", "ApiCenterApi", "API…

microsoft/skills · 97 tokens

azure-mgmt-mongodbatlas-dotnet

Manage MongoDB Atlas Organizations as Azure ARM resources using Azure.ResourceManager.MongoDBAtlas SDK. Use when creating, updating, listing, or deleting MongoDB Atlas organizations through Azure Marketplace integration. This SDK manages the Azure-side organization resource, not Atlas clusters/databases directly.

microsoft/skills · 64 tokens