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 agentmods add commands/pashki975/thm-claude-kit/steggit 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/steg)<a href="https://agentmods.dev/commands/pashki975/thm-claude-kit/steg"><img src="https://agentmods.dev/badge/commands/pashki975/thm-claude-kit/steg.svg" alt="Measured on agentmods" 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 | $0.00404 |
| Opus 5 | $0.00007 | $0.00202 |
| Sonnet 5 | $0.00003 | $0.00081 |
| Haiku 4.5 | $0.00001 | $0.00040 |
Grade A, and why
steg 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 5d 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
Analyze the file(s): $ARGUMENTS (a path, or a directory of downloaded files)
Use when a room hands you a file to download and the flag is hidden in or behind it. Work cheap-to-slow. The flag is the goal — most of these need no shell.
1. Identify
- file — real type regardless of extension
- exiftool — metadata: authors, comments, GPS, flags in fields
- strings -n 6 — grep for flag{, THM{, base64, URLs
2. Embedded / appended data
- binwalk ; binwalk -e (or foremost )
- try unzip/7z x on an image (polyglots)
3. Image steganography
- steghide info ; steghide extract -sf (blank passphrase + room passwords)
- zsteg (PNG/BMP LSB) ; stegseek (brute steghide w/ rockyou)
- stegsolve for visual LSB / colour-plane inspection
4. Audio / docs
- Audio: spectrogram in Audacity/sonic-visualiser (flags drawn there)
- PDFs: pdfdetach -list, pdf-parser ; Office: unzip and read XML/macros
5. Protected/hashed
- Hand off to /crack (zip2john etc.)
6. Summarize
Say what the file was, what was hidden, how you got it. If you recovered the flag, stop. Record in notes.md. Note: steg is mostly beginner/jeopardy-CTF — on a normal box, don't rabbit-hole where straightforward enumeration is intended.
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.
- 5d ago First seen · 35 lines · 14 tokens per session scan A 2693f75a8548
steg is a command published in the GitHub repository pashki975/thm-claude-kit (2 stars, last pushed 22d ago), licensed MIT. It adds 14 tokens to every session and 404 once invoked, about $0.0001 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 2026-08-31.
Other commands, from other repositories
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.
calibrate-voice
Run an empirical calibration round: regenerate the 8-brief matrix, measure, collect owner ratings, codify every reaction into a logged pack change.
build
Build it — phase by phase with code review and your approval at each gate.
comments
Show cubic's review comments on a pull request.