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 meltedinhex/analyst-ai-pack --skill hunting-suspicious-powershell-executiongit clone --depth 1 https://github.com/meltedinhex/analyst-ai-packWrote 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/meltedinhex/analyst-ai-pack/hunting-suspicious-powershell-execution)<a href="https://agentmods.dev/skills/meltedinhex/analyst-ai-pack/hunting-suspicious-powershell-execution"><img src="https://agentmods.dev/badge/skills/meltedinhex/analyst-ai-pack/hunting-suspicious-powershell-execution/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/meltedinhex/analyst-ai-pack/hunting-suspicious-powershell-execution"><img src="https://agentmods.dev/badge/skills/meltedinhex/analyst-ai-pack/hunting-suspicious-powershell-execution.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 | $0.00079 | $0.00800 |
| Opus 5 | $0.00039 | $0.00400 |
| Sonnet 5 | $0.00016 | $0.00160 |
| Haiku 4.5 | $0.00008 | $0.00080 |
Grade A, and why
hunting-suspicious-powershell-execution 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 6d 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.
How it starts
The opening of the file, as written. The whole thing — 95 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hunting Suspicious PowerShell Execution
When to Use
- You have PowerShell script-block (Event ID 4104) and/or module logging and want to hunt abuse.
- You are testing a hypothesis that attackers use encoded commands, download cradles, or AMSI/logging bypasses.
- You need to score and decode a large volume of PowerShell events into a triage queue.
Do not use simple keyword blocklists as the sole method — admins legitimately use many of the same cmdlets; score combinations and decode before judging.
Prerequisites
- Script-block logging enabled (Event ID 4104) and exported to a queryable store.
- A baseline of normal administrative PowerShell for your environment.
Workflow
Step 1: Collect script-block events
Pull EID 4104 script-block text (and EID 4103 module logging) over the hunt window. Reassemble multi-part script blocks before analysis.
Step 2: Score suspicious constructs
Weight combinations rather than single keywords: -EncodedCommand, IEX/Invoke-Expression
with Net.WebClient/Invoke-WebRequest, FromBase64String, -w hidden -nop, AMSI strings,
and [Ref].Assembly reflection.
python scripts/analyst.py score events.json
Step 3: Decode payloads
Base64-decode -EncodedCommand (UTF-16LE) and recover layered encodings to read the real
intent (download URLs, in-memory loaders).
Step 4: Baseline and pivot
Compare against normal admin activity; for survivors, pull the parent process, user, host, and network connections to confirm.
Step 5: Confirm and operationalize
Validate true positives, escalate to IR, and convert the scoring logic into a detection rule.
Validation
- High-score events decode to coherent malicious intent, not benign admin scripts.
- Multi-part script blocks are reassembled before scoring (no truncated false negatives).
- Confirmed cases tie to a parent process and user consistent with intrusion.
Pitfalls
- Flagging single keywords (
IEXalone) and drowning in admin noise. - Forgetting
-EncodedCommandis UTF-16LE, producing garbled decodes. - Ignoring AMSI/logging-bypass attempts that precede the real payload.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 6d ago First seen · 95 lines · 79 tokens per session scan A 6a9a6799d7ff
hunting-suspicious-powershell-execution is a skill published in the GitHub repository meltedinhex/analyst-ai-pack (22 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 79 tokens to every session and 800 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 2026-09-03.
Other skills, from other repositories
detection-sigma
Generic detection rule creation and management using Sigma, the universal SIEM rule format. Sigma provides vendor-agnostic detection logic for log analysis across multiple SIEM platforms. Use when: (1) Creating detection rules for security monitoring, (2) Converting rules between SIEM platforms (Splunk, Elastic…
analyzing-cobalt-strike-beacon-configuration
Extract and analyze Cobalt Strike beacon configuration from PE files and memory dumps to identify C2 infrastructure, malleable profiles, and operator tradecraft.
analyzing-golang-malware-with-ghidra
Reverse engineer Go-compiled malware in Ghidra by parsing Go buildinfo and pclntab structures, recovering stripped/obfuscated function names (e.g. via GoResolver), and extracting embedded module/dependency strings and types from Go binaries. Use when analyzing a Go-language malware sample, deobfuscating a…
analyzing-malware-persistence-with-autoruns
Use Sysinternals Autoruns to systematically enumerate and analyze malware persistence mechanisms across Windows registry run keys, scheduled tasks, services, drivers, and startup locations. Use when hunting for persistence during Windows incident response, triaging a compromised endpoint, or validating that malware…
analyzing-malicious-pdf-with-peepdf
Perform static analysis of malicious PDF documents using peepdf, pdfid, and pdf-parser to extract embedded JavaScript, shellcode, and suspicious objects. Use when triaging a suspicious PDF attachment from a phishing email, analyzing a PDF-based exploit document, or building detection signatures for weaponized PDF…
analyzing-powershell-script-block-logging
Parse Windows PowerShell Script Block Logs (Event ID 4104) from EVTX files to detect obfuscated commands, encoded payloads, and living-off-the-land techniques. Uses python-evtx to extract and reconstruct multi-block scripts, applies entropy analysis and pattern matching for Base64-encoded commands, Invoke-Expression…