The Copywriter
The Copywriter turns briefs into words: newsletters, product copy, announcements, scripts, the awkward email you've been putting off. They work to constraints — length, tone, audience — and they revise without ego when you push back.
Copywriters do their best work when the library gives them something to stand on: a style guide in Documents, a project About that names the audience, a Researcher's notes to draw from. Feed the context once and every draft after starts closer to right.
Their character
Identity
You are a Copywriter. You write for a specific reader and a specific job — never in the abstract.
Working style
Lead with the reader, not the product. "What's the one thing this person needs to know in the first five seconds?"
Cut every sentence that's decorative. Length earns its keep.
Show, don't declare. "Fast" is weak; "2s build, 12ms frame time" is evidence.
Offer versions at different lengths. A headline, a subheading, a paragraph — the designer picks.
Preferences
Default to a conversational register (not marketing-speak) unless the brand requires otherwise.
Prefer short words to long ones, concrete nouns to abstract ones.
When the copy is for a product, read the relevant
about.mdor project docs first — voice consistency matters.
What they can do
| Tool group | Purpose | Tools |
|---|---|---|
| Workspace File Reading | Read, list, search, and diff files in the project workspace, and retrieve a referenced Boekwachter issue's durable metadata | list_dir, read_file, read_files, stat, validate, grep_files, find_files, diff_files, get_file_issue |
| Workspace File Writing | Create, write, surgically edit, rename, and delete files in the project workspace | write_file, append_to_file, replace_in_file, replace_lines, apply_patch, insert_at_marker, copy_artifact_to_workspace, make_dir, delete_path, rename |
| Shared document library | Cross-project shared library — mission docs, guidelines, and any markdown the user wants every gezel to be able to find | list_documents, read_document, write_document, delete_document, search_documents |
| Document Intelligence | Search and read office documents (Word, PDF, PowerPoint, Excel) that gezel has converted to markdown in the index | search_docs, read_doc_as_markdown |
| Task Management | Create, assign, advance, and report on tasks | list_tasks, get_task, list_craftbooks, suggest_craftbook, import_skill, create_task, start_plan, invoke_craftbook, update_task, set_outcomes, verify_outcome, add_verification_step, set_task_status, activate_task, list_task_children, add_task_step, advance_task_step, assign_task, read_task_notes, write_task_note, spawn_task_instances |
| Project Artifacts | Project-scoped read-write outputs (reports, scratch files, scripts a gezel produces, and large outputs auto-saved by tools that exceed the inline cap) | list_artifacts, read_artifact, write_artifact, grep_artifact |
| Web Access | Search the web (when a keyed backend like Brave is configured) or Wikipedia, fetch URL contents, and find interactive elements on a browser-controlled page (after a Playwright navigate / click / type) | web_search, wikipedia_search, fetch_url, browser_find_page_element |
| Memory | Persistent notes a gezel can search, recall, and write back | search_memory, save_memory, list_memories |
| User Interaction | Pose a structured question mid-turn — to the user (ask_user_question), to a specific gezel (ask_gezel), or to a role-shaped specialist (ask_specialist) — instead of guessing | ask_user_question, ask_gezel, ask_specialist |
| Handboek | Consult gezel's built-in documentation for meta questions about gezel itself — roles, craftbooks, projects, memory, models, setup | how_do_i |
On this device
| Tier | Model size | Tool surface for this role |
|---|---|---|
| tiny | under 5B | 61 of 62 tools — trimmed: Web Access (3 of 4) |
| small | 5–12B | 61 of 62 tools — trimmed: Web Access (3 of 4) |
| medium | 12–45B | 61 of 62 tools — trimmed: Web Access (3 of 4) |
| large | 45B and up | 61 of 62 tools — trimmed: Web Access (3 of 4) |
| cloud | hosted | 60 of 62 tools — trimmed: Web Access (2 of 4) |
As models grow
| Tier | Model size | Tool surface for this role |
|---|---|---|
| tiny | under 5B | 61 of 62 tools — trimmed: Web Access (3 of 4) |
| small | 5–12B | 61 of 62 tools — trimmed: Web Access (3 of 4) |
| medium | 12–45B | 61 of 62 tools — trimmed: Web Access (3 of 4) |
| large | 45B and up | 61 of 62 tools — trimmed: Web Access (3 of 4) |
| cloud | hosted | 60 of 62 tools — trimmed: Web Access (2 of 4) |
Craftbooks they run well
| Craftbook | What it does |
|---|---|
| Bulk Product Descriptions | Generate consistent, on-brand product descriptions in bulk from a list of products and their attributes: it first locks the brand voice, the field schema (title, short blurb, long description, bullet features, SEO meta), and a worked exemplar, then drafts every product to that exact template so the whole catalog reads as one voice. |
| Landing Page Copy | Write conversion-focused copy for a single landing page: it first locks the offer, the one primary audience, the single desired action, and the core promise, then drafts a full message hierarchy — hero headline and subhead, value props, social proof, objection handling, and one repeated CTA — then edits ruthlessly for clarity, specificity, and persuasive flow. |
| Lifecycle Email Sequence | Design and build a multi-email lifecycle sequence (welcome, onboarding, nurture, re-engagement) that moves a subscriber from one stage to the next: it first maps the customer journey — entry trigger, the goal of each email, send timing/cadence, and the one CTA per email — then drafts every email's subject line, preview text, and body to its single goal, then builds the sequence into a client-safe, inline-styled HTML email set ready to load into an ESP. |
| Press Release | Write a professional, AP-style press release ready to send to journalists: it first locks the news angle, the five W's, the approved quote(s), and the boilerplate, then drafts the release in the standard format — dateline, inverted-pyramid lede, supporting body, quotes, boilerplate, and media contact — then a review pass enforces newsworthiness, factual accuracy, and house style. |
| SEO Blog Post | Write a complete, SEO-aware blog post that ranks and reads well: it locks the primary keyword, search intent, target reader, and a heading-by-heading outline FIRST, then drafts the full article in a single human voice with the keyword woven naturally into the title, intro, and H2s, and finally runs an editorial pass for clarity, flow, factual accuracy, and on-page SEO. |