new-agent

A command for creating a new Claude Code subagent from a guided interview. It produces an agent file describing the agent's purpose, scope, tools, questions, common mistakes, debugging process, and output format.

In plain words
What is it for?
Use it to scaffold agents for a particular language, platform, framework, or technical job, then preview and confirm the generated configuration.
Why use it?
It helps define a specialized agent clearly before adding it to a project, reducing vague roles and missing operating rules.

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/ackeskin/contexture/new-agent
Clone the repo
git clone --depth 1 https://github.com/AcKeskin/contexture
Per session 54 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 301 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 $0.00054 $0.00301
Opus 5 $0.00027 $0.00151
Sonnet 5 $0.00011 $0.00060
Haiku 4.5 $0.00005 $0.00030

Measured 2d ago against content hash add7bf8c494c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

new-agent 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.

commands/new-agent.md · 18 lines

What it actually says

Run the new-agent skill for the current context.

Any text after /new-agent is a hint that biases the interview but never enough to skip it. Examples:

  • /new-agent — full interview from scratch.
  • /new-agent metal video source — hint pre-fills the domain question; skill still confirms.
  • /new-agent rust async runtime expert — hint pre-fills name and scope; skill still confirms every required field.

The skill walks through nine stages: name + slug, job-to-be-done, scope (language/platform/framework), tool allowlist, model choice, pre-flight questions the agent must always ask, anti-patterns (specific landmines), debugging workflow, output contract. It then composes the full agent file with all sections, shows a preview, asks for confirmation, writes to agents/<name>.md, and reminds the user that bootstrap propagates it to ~/.claude/agents/.

Mode A only — never auto-fires. The whole point is to push the user to articulate what makes this agent worth having beyond a generic coder.

See ~/.claude/skills/new-agent/SKILL.md for the full procedure.

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. 2d ago First seen · 18 lines · 54 tokens per session scan A add7bf8c494c

Subscribe to this mod's changes

new-agent is a command published in the GitHub repository AcKeskin/contexture (2 stars, last pushed 1mo ago), licensed MIT. It adds 54 tokens to every session and 301 once invoked, about $0.0003 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.