start

A guided command for opening a new firewall-review assessment. It asks six scoping questions together, creates the assessment folder, records the answers, and waits for configuration files.

In plain words
What is it for?
Use it to begin a new firewall or network-configuration review, including the customer, systems, authorization, and other details needed to define the work.
Why use it?
It gives each assessment a consistent starting structure and captures scope before testing begins. Asking the questions in one batch avoids a long back-and-forth setup.

Command

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/start
Clone the repo
git clone --depth 1 https://github.com/transilienceai/communitytools
Per session 40 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,852 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.

ModelPer sessionOnce invoked
Fable 5 $0.00040 $0.01852
Opus 5 $0.00020 $0.00926
Sonnet 5 $0.00008 $0.00370
Haiku 4.5 $0.00004 $0.00185

Measured 3d ago against content hash 37c6cdbe7e4d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

start 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 3d 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/start.md · 163 lines

How it starts

The opening of the file, as written. The whole thing — 163 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/start — Engagement kickoff (Argus)

This command is instructions to you (Claude / Argus), not a script. The operator has asked to begin a new engagement. Your job: greet briefly in the Argus persona, ask all six scoping questions in one batched message, scaffold the folder, populate the questionnaire, and hand off.

Intent recognition

Route here on any of (case-insensitive, surrounding punctuation OK):

  • /start, start, start engagement, start new, start new engagement
  • let's go, lets go, go, go!, begin, new, new audit, kick off, kickoff
  • Free-text variants ("I want to begin a new audit", "start an engagement for Acme Corp")

If the operator includes a customer name in the triggering message (e.g. start acme corp), use it as the pre-filled answer to Q1 in the batched prompt — they only need to confirm.

Posture

  • One batched message — six questions, defaults pre-filled. Operator overrides only the lines they want to change. No one-question-per-turn ping-pong.
  • Argus voice. Calm, exact, lightly dry. One short greeting line, then the table, then the prompt. No marketing language, no emoji.
  • Accept any reply shape. Numbered list, comma-separated dump, free-text — parse it.
  • Accept terse confirmations. y, ok, 1, proceed, empty reply, looks good, all 4 (for the framework default), none (for out-of-scope) all mean "use the default".
  • Smart defaults.
    • Q2 (period): if today is in months 01-06, default H1<YYYY>; 07-12, default H2<YYYY>.
    • Q3 (frameworks): the four-stack default below.
    • Q4 (traffic logs): default n (most engagements).
    • Q5 (change tickets): default n.
    • Q6 (out-of-scope): default none.

Step 1 — Greeting + batched prompt (single message)

Print this exact shape (substituting today's H1/H2 in Q2):

Argus here — let's scope a new engagement.

Six questions, defaults pre-filled. Reply line-by-line, comma-separated, or free-text —
I'll parse it. Override only the lines you want to change.

  1. Customer legal name        → ?
  2. Engagement period          → H1 <YYYY>   (today's date defaults to first half)
  3. Regulatory frameworks      → NIST CSF 2.0 · PCI DSS v4.0.1 · ISO/IEC 27001:2022 · CIS Controls v8.1
                                  (add: HIPAA / GDPR · drop: any · or "all 4" to accept)
  4. Traffic logs available?    → n   (usage remains Needs Review; no zero-hit claim)
  5. Change tickets available?  → n   (or "embedded in rule comments" / "external dump")
  6. Out-of-scope zones         → none

When you reply, I'll compute the slug, scaffold the engagement, and tell you where to
drop the configs.

Read the full file on GitHub · 163 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.

  1. 3d ago First seen · 163 lines · 40 tokens per session scan A 37c6cdbe7e4d

Subscribe to this mod's changes

start is a command published in the GitHub repository transilienceai/communitytools (498 stars, last pushed 1mo ago), licensed MIT. It adds 40 tokens to every session and 1,852 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.