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/ogrodev/fsociety/ty-classifygit clone --depth 1 https://github.com/ogrodev/fsocietyWrote 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/ogrodev/fsociety/ty-classify)<a href="https://agentmods.dev/commands/ogrodev/fsociety/ty-classify"><img src="https://agentmods.dev/badge/commands/ogrodev/fsociety/ty-classify.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 | $0.00014 | $0.00936 |
| Opus 5 | $0.00007 | $0.00468 |
| Sonnet 5 | $0.00003 | $0.00187 |
| Haiku 4.5 | $0.00001 | $0.00094 |
Grade A, and why
ty-classify scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 109 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are classifying a discovered data source for the tyrell data exfiltration plugin.
Step 1 — Retrieve Source Details
Look up the source in the source-tracker to display its current metadata:
node "${CLAUDE_PLUGIN_ROOT}/scripts/source-tracker.js" list --id $ARGUMENTS
Display the full source record to the operator:
- Source ID, type, URL/host
- Discovery method and timestamp
- Any notes recorded at discovery time
- Current classification status (likely
unclassified)
Step 2 — Gather Classification Input
Ask the operator a series of classification questions. Ask them all at once to avoid multiple round-trips:
Ask: "Please classify this source across the following dimensions:
1. Credibility — How confident are you that this source contains real, non-fabricated data?
unverified— No sample seen, source claim onlylow— Sample seen but unconfirmed (could be synthetic)medium— Sample partially verified against known-good datahigh— Sample verified against live target or known breachconfirmed— Fully verified, matches known breach event
2. Freshness — How recent is the data?
unknown— Date not determinablestale— Older than 2 yearsaged— 6 months to 2 years oldrecent— Less than 6 months oldfresh— Less than 30 days old
3. Access Method — How is the data accessible?
open— Publicly accessible without authenticationregistration— Requires free account/registrationpurchase— Requires payment (specify amount in notes)trade— Available via data exchange/tradeexploit— Requires active exploitation to retrievephysical— Requires physical access
4. Data Sensitivity — What is the highest-sensitivity data type present?
public— Already public informationpii— Personal identifiable information (names, emails, phones)credentials— Username/password pairsfinancial— Payment card or banking datahealth— Medical or health recordsgovernment— Government or law enforcement data
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.
- today First seen · 109 lines · 14 tokens per session scan A c9925d77a38a
ty-classify is a command published in the GitHub repository ogrodev/fsociety (20 stars, last pushed 5mo ago), licensed MIT. It adds 14 tokens to every session and 936 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 2026-09-03.
Other commands, from other repositories
pentest
Activate pentest mode — displays ASCII art, configures session isolation, collects engagement scope, then OWNS the engagement: pre-flight, recon, planning (via the pentester-orchestrator planner), executor dispatch, a time-budget quota loop, aggregation, and report generation.
pentest-attacks
Define the attack profile for an engagement — select which attack categories and skills to use. Saves to .pentest-attacks.json. If run before /pentest:pentest, the orchestrator will respect the selection. If run standalone, does not launch a pentest.
pentest-kali
Connect to a Metasploit-Kali Server (MKS) REST API — verifies connectivity, discovers available Kali tools, and configures agents to prefer MKS endpoints over local Bash equivalents.
pentest-scope
Define or update engagement scope — saves scope to disk without launching a pentest. Can be run before or during an engagement. If a pentest is active and the target changes drastically, warns the operator and suggests a new engagement.
pentest-exit
Close pentest session — summarizes findings, ensures outputs are saved, lifts isolation, and prompts for /clear.
bb-ad
Active Directory enumeration and attack techniques. Includes LDAP enumeration, Kerberos attacks (Kerberoasting, AS-REP Roasting), SMB attacks, and domain privilege escalation. Use this when targeting Windows domain environments.