copy-edit
Copy-edits your spoken transcript or rough draft into a finished post that still sounds like you.
Skill by ptrandev
Chalk Labs
Need help deploying copy-edit?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/ptrandev-copy-edit
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: writing
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/copy-edit
What's inside
- Skill name
copy-edit- Description
- Copy-edits your spoken transcript or rough draft into a finished post that still sounds like you. Restructures freely, cuts the spoken filler, adds hyperlinks it verified live, and never adds a claim you did not make. Use for "copy edit this", "I recorded this, turn it into a post", "clean up this transcript", or "edit my draft".
- License
- MIT
- Instructions
- 213 lines in SKILL.md · view on GitHub
Before installing
- Instructions call the network
Files (1)
- SKILL.md
SKILL.md (first lines)
# copy-edit
You are the copy editor. The author is the writer. **Never** add a claim, a number, a reason,
or an opinion the author did not say. Every sentence in the finished draft traces back to a
line in the source.
The source is speech. It is disorganized, repetitive, and full of restarts. It is also
already correct about the subject. Your job is presentation, not content.
## Input
| Invocation carries | Source |
|---|---|
| Pasted transcript text | That text. |
| A file path | That file. |
| Both | The file is the source. The message text is instruction. |
**Never** guess the output path. Ask for it in Phase 8, after the title exists.
## Phase 1: Inventory the spine
Before you edit anything, write a scratch inventory of the source:
1. Every claim, and the evidence the author gave for it.
2. Every number, with its unit, its hedge, and what it measures.
3. Every named product, model, company, and comparison.
4. Every caveat, condition, and exception ("one thing to watch", "in our case").
5. Every recommendation, with the situation it applies to.
This inventory is the checklist for Phase 8. Nothing on it disappears without a line in the
report.
## Phase 2: Voice fingerprint
Derive the voice from the source, not from a generic blog voice. Record:
- Person and address: "I", "we", "you".
- Contractions: present or absent.
- Sentence rhythm: where the author runs long, where the author snaps short.
- Signature moves: "reach for", "as a rule of thumb", "one thing to watch".
- Stance words the author grades things with: "amazing", "excellent", "worth it".
- How the author hedges: "roughly", "about", "typically", "in our case".
Read one earlier post by the author when one sits in the output directory or the author links
one. Use it for the fingerprint only.
**Reword freely inside the fingerprint.** Write the clearest version of each point, in words
that sound like the author speaking out loud. **Never** upgrade the register. No "leverage", "utilize",
"furthermore", "moreover". A sentence that sounds smarter than the author is a failed edit.
## Phase 3: Restructure
You have full latitude. Reorder sections, rewrite headings, promote a buried point into its
own section, and merge two sections that argue one thing.
- A heading states the claim, not the topic. "Reach for a larger model at lower reasoning"
beats "Model selection".
- Lead each section with its claim. Put the evidence after it.
- Move a spoken digression under the claim it supports.
- Keep the author's order when it already works. Reorder to fix a forward reference, or to
reunite an argument the author split across two places.
- **Never** merge two claims into one sentence. **Never** split one claim across two sections
to pad the structure.
- Cut a digression that supports nothing, and name it in the report.
## Phase 4: Line edit
Cut, because speech carries it and text does not:
- Verbal filler: "so", "basically", "you know", "kind of", "I mean", "right", "actually".
- Throat-clearing openers: "So what I did here was", "The thing is", "What I'll say is".
- Restarts and repeated words from the recording.
- The same point made three times. Keep the sharpest phrasing, once.
Fix:
- Spoken run-ons. Split at the "and" that joins two complete ideas.
- Dangling referents. "That", "this", and "it" with no antecedent get the thing named.
- Tense drift inside one passage.
Keep:
- Direct address to the reader.
- Every number exactly as the author said it, hedge attached.
- The author's stance words, including the informal ones.
## Phase 5: Links
Add hyperlinks so the reader can get context on what they are reading.
Earns a link, on first mention only:
- A product, company, or project.
- A specific model, library, or tool.
- A documented feature or concept the reader does not necessarily know: reasoning effort, prompt caching,
minimum cacheable length.
- A standard, spec, or public source behind a number.
**Never** link: a heading, a term already linked in the post, a common word, or the author's
own claim about the author's own data.
Anchor text is the name of the thing. **Never** "here", "this doc", or "click here". **Never**
wrap a whole clause.
Density: at most three links in one paragraph. **Never** two adjacent.
Verify every candidate before it enters the draft:
1. Find the canonical URL. Prefer the vendor's own documentation over a blog or an aggregator.
2. Fetch it with `WebFetch`. `curl -sIL -o /dev/null -w '%{http_code} %{url_effective}\n' <url>`
is a fast pre-check, and a sandbox can block it, so `WebFetch` is the one that decides.
3. Confirm that the page is about the anchor text. A redirect to a generic index means the
deep page is gone.
4. Drop any candidate that fails, and list it in the report.
Run the whole batch of fetches in parallel, in one message. Re-verify links the source draft
already contains. Report a dead one. **Never** delete an author's link silently.
**Never** insert a URL you did not fetch in this run.From ptrandev/agent-skills, MIT.
More from ptrandev/agent-skills
- 0.592score
pointsgpt-support
ptrandev
Triages the PointsGPT #pointsgpt-support Slack alerts.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
review-pr
ptrandev
Reviews GitHub PRs where you are the requested reviewer, applying Phillip's engineering bar.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
ui-walkthrough
ptrandev
Walks a PR's UI changes and demonstrates its behavior changes in a real browser, judges what it sees, and posts the screenshots and video to GitHub as the…
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
babysit-prs
ptrandev
Triages unresolved review threads on open PRs that you authored, fixes the safe ones, replies with the fixing commit, and leaves anything ambiguous open with…
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
full-send
ptrandev
End-to-end feature workflow: take a Linear ticket or a raw idea, implement it, self-review, open a PR, address every bot review thread, then capture UI and…
SkillAFreshness: A★ 1Claude Code +9 - 0.583score
launch-summary
ptrandev
Summarizes what shipped across the Atllas repos over a daily or weekly window, written for non-developers.
SkillAFreshness: A★ 1Claude Code +9 - 0.580score
phillip
ptrandev
Self-review of the current uncommitted change against Phillip's engineering bar, before it becomes a PR.
SkillAFreshness: A★ 1Claude Code +9 - 0.579score
delegate
ptrandev
Tells the agent to hand work to subagents and stay the orchestrator.
SkillAFreshness: A★ 1Claude Code +9 - 0.552score
claude
ptrandev
Calls the Claude Code CLI for an independent review, adversarial challenge, or codebase consultation.
SkillBFreshness: B★ 1Claude Code +9 - 0.552score
plain-english
ptrandev
Extracts the signal from bloated or evasive text: the bottom line, every claim with its strength marked, what it does not say, and what it implies without…
SkillBFreshness: B★ 1Claude Code +9 - 0.552score
mockup
ptrandev
Builds one self-contained HTML file that looks like the real product and walks a proposed change through every state, interactively, before the change is built.
SkillBFreshness: B★ 1Claude Code +9 - 0.547score
skill-edit
ptrandev
Rewrites one skill file in this repository to the house style in AGENTS.md, without losing an instruction.
SkillBFreshness: B★ 1Claude Code +9
Signals
- Stars
- 1
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Sep 4, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/ptrandev/agent-skills
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from ptrandev
- 0.592score
pointsgpt-support
ptrandev
Triages the PointsGPT #pointsgpt-support Slack alerts.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
review-pr
ptrandev
Reviews GitHub PRs where you are the requested reviewer, applying Phillip's engineering bar.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
ui-walkthrough
ptrandev
Walks a PR's UI changes and demonstrates its behavior changes in a real browser, judges what it sees, and posts the screenshots and video to GitHub as the…
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
babysit-prs
ptrandev
Triages unresolved review threads on open PRs that you authored, fixes the safe ones, replies with the fixing commit, and leaves anything ambiguous open with…
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
full-send
ptrandev
End-to-end feature workflow: take a Linear ticket or a raw idea, implement it, self-review, open a PR, address every bot review thread, then capture UI and…
SkillAFreshness: A★ 1Claude Code +9 - 0.583score
launch-summary
ptrandev
Summarizes what shipped across the Atllas repos over a daily or weekly window, written for non-developers.
SkillAFreshness: A★ 1Claude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get copy-edit running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
copy-edit skill. Agent Mart. https://agentmart.dev/skills/ptrandev-copy-edit (last updated 2026-10-24)