A command that runs a security-testing pipeline for the newest assessment engagement. It reads configuration files, sends testing tasks to specialist agents, validates the results, and prepares findings for review.

Use it to start detection and validation for firewall or other security configuration assessments and produce a findings file for the review step.

It removes the manual work of finding the right engagement, classifying vendor configuration files, and coordinating the testing sequence. It also stops when the required configuration files are missing.

launch — Command by transilienceai

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 commands/transilienceai/communitytools/launch

Clone the repo

git clone --depth 1 https://github.com/transilienceai/communitytools

Per session 32 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 2,624 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.

Model	Per session	Once invoked
Fable 5	$0.00032	$0.02624
Opus 5	$0.00016	$0.01312
Sonnet 5	$0.00006	$0.00525
Haiku 4.5	$0.00003	$0.00262

Measured 2d ago against content hash 761f81d7716b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

launch 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 2d 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/firewall-review/reference/commands/launch.md · 121 lines

How it starts

The opening of the file, as written. The whole thing — 121 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/launch — Run the detection + validation pipeline

No arguments. Claude infers the engagement directory from the most recently scaffolded engagements/*/ folder, OR asks the operator which engagement.

What to do

Step 1 — Locate the engagement and classify inputs

ls engagements/ and pick the most recent folder OR ask the operator if there are multiple open engagements. Accept both naming conventions: new engagements/<slug>_<date>/ AND legacy engagements/<date>-<slug>/.
Confirm there are files under Pre-requisites/. If the folder is empty (no top-level files AND no legacy vendor sub-folders containing configs), STOP and remind the operator: "No configs found. Drop files into Pre-requisites/ first — any vendor, any filename."
Walk Pre-requisites/ flat. The new layout has no per-vendor sub-folders; every config file sits at the top of Pre-requisites/. For each file (skipping README.md, scoping-questionnaire.yaml, and .DS_Store), call fwrr.parsers._sniff.detect_vendor(path) and report the classification to the operator before parsing:
```
Input classification:
  <customer-slug>-va1-fgtfwa.conf    → fortigate (conf 0.95)
  acme-panfw-east.xml                → pan-os-xml (conf 1.00)
  nsg-prod-east.json                 → azure-nsg (conf 0.95)
  edge-unknown.txt                   → unknown (conf 0.00)  [skipped]
```
If any file classifies as unknown or drops below confidence 0.75, warn the operator and give a fix hint: "rename the file to include a vendor hint (e.g. *-fortigate.conf, *-asa.cfg, *-panos.xml) OR drop it into the legacy per-vendor layout (Pre-requisites/<vendor>/)." Skip the file and continue — don't block the run.
Backward compatibility. If the engagement still uses the legacy layout with Pre-requisites/<vendor>/ sub-folders (e.g. engagements/YYYY-MM-DD-<customer-slug>/), the launcher walks those sub-folders automatically and trusts the folder name without running the sniff — old engagements keep working unchanged. Report which layout you detected in the summary line.

Read the full file on GitHub · 121 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.

2d ago First seen · 121 lines · 32 tokens per session scan A 761f81d7716b

Subscribe to this mod's changes

launch is a command published in the GitHub repository transilienceai/communitytools (498 stars, last pushed 1mo ago), licensed MIT. It adds 32 tokens to every session and 2,624 once invoked, about $0.0002 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.

Code review Security Testing bounty-hunters hackerone pentest pentest-tool security-research

launch

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

/launch — Run the detection + validation pipeline

What to do

Step 1 — Locate the engagement and classify inputs

What this file has done since we first saw it

Other commands, from other repositories

checklist

clarify

specify

analyze

constitution

converge