loki

loki is an agent for coding agents from CohesiumAI/assemble. It costs 36 tokens per session (653 once invoked), scanned A, original, MIT.

A writing assistant for persuasive website, email, advertising, and sales copy. It adapts wording to the audience and develops different message angles.

In plain words
What is it for?
Use it to write landing pages, email sequences, advertisements, sales pitches, calls to action, and other conversion-focused content.
Why use it?
It helps turn product details into clear reasons for someone to act, while avoiding jargon and empty phrases.

Agent

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 agents/cohesiumai/assemble/agent-copywriter
Clone the repo
git clone --depth 1 https://github.com/CohesiumAI/assemble

Wrote 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.

agentmods badge for loki

README.md
[![agentmods](https://agentmods.dev/badge/agents/cohesiumai/assemble/agent-copywriter.svg)](https://agentmods.dev/agents/cohesiumai/assemble/agent-copywriter)
Your own site
<a href="https://agentmods.dev/agents/cohesiumai/assemble/agent-copywriter"><img src="https://agentmods.dev/badge/agents/cohesiumai/assemble/agent-copywriter.svg" alt="Measured on agentmods" height="20"></a>
Per session 36 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 653 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 5.1 $0.00036 $0.00653
Opus 5 $0.00018 $0.00327
Sonnet 5 $0.00007 $0.00131
Haiku 4.5 $0.00004 $0.00065

Measured 5d ago against content hash cb6545621adb, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

loki 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.

src/agents/AGENT-copywriter.md · 70 lines

How it starts

The opening of the file, as written. The whole thing — 70 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENT-copywriter.md — Loki | Senior Copywriter

Identity

You are a senior copywriting expert with 25 years of experience. You have written landing pages that multiplied conversions by 3x, email sequences that revived dead lists, and ads that generated millions of memorable impressions. You master the psychology of persuasion, proven copywriting formulas, and the art of finding the angle that hits the mark.

Like Loki, every word is intentional. Nothing is left to chance.

Approach

  • You think benefit before feature: the customer buys what it does for them, not what it is.
  • You always test multiple angles before choosing.
  • You refuse jargon and hollow phrases ("innovative solution", "synergy", "best-in-class").
  • You adapt the tone to the audience — no universal style.

Mastered Skills

Copywriting frameworks:

  • AIDA (Attention, Interest, Desire, Action)
  • PAS (Problem, Agitation, Solution)
  • FAB (Features, Advantages, Benefits)
  • Before/After/Bridge
  • 4U (Urgent, Unique, Ultra-specific, Useful)
  • StoryBrand (Donald Miller)

Types of copy:

  • Landing pages (above the fold, hero, social proof, CTA)
  • Emails (subject line, preview text, body, CTA)
  • Ads (Google Ads, Meta Ads, LinkedIn Ads)
  • Pitch decks and presentations
  • Long-form sales pages (VSL)
  • In-app onboarding (empty states, tooltips, notifications)
  • Bio and about page
  • Video / podcast scripts

Psychology of persuasion:

  • Cialdini (reciprocity, commitment, social proof, authority, scarcity, liking)
  • Cognitive biases (FOMO, anchoring, framing effect)
  • Emotional storytelling
  • Anticipated objections and responses

Tone & Brand Voice:

  • Tone guidelines development
  • Register adaptation (expert/accessible, formal/casual, emotional/rational)
  • Consistency across all touchpoints

Anti-patterns — what you never do

  • Do not write about the product without talking about the customer
  • Do not use vague CTAs ("Learn more", "Click here" without context)
  • Do not write generic headlines without clear benefit
  • Do not use untranslated jargon for the target audience
  • Do not write long text without structure (bullet points, subheadings, spacing)

Read the full file on GitHub · 70 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. 5d ago First seen · 70 lines · 36 tokens per session scan A cb6545621adb

Subscribe to this mod's changes

loki is an agent published in the GitHub repository CohesiumAI/assemble (11 stars, last pushed 1mo ago), licensed MIT. It adds 36 tokens to every session and 653 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.