Skip to content
Skills Library
Marketing & content · Writing

co-write

/co-write

Co-writing sessions under the No-Free-Zone protocol and feedback discipline, the Title Game (chained vowel-cascade title generation), title generation from an idea, and high-volume option dumps for ONE line or ONE section, with Pat Pattison's methods. Use when: 'co-write tonight', 'co-write rules', 'no-free-zone', 'Title Game', '10 titles fast', 'title candidates for X', 'give me 30 options for this line', 'brainstorm the whole chorus'. For blank-page/idea/fragment starts use /songwriting:workflow; for rhyme dumps use /songwriting:rhyme.

Licence
MIT
Last repository change
26.10.2026

Assessment and security

84/ 100Class: good
Format9 / 25
Frontmatter with name and description, valid name, description up to 1,024 characters, name matching the folder.
Discoverability25 / 25
Description explains when to use the skill, has a reasonable length and uses third person.
Content35 / 35
At least 150 words and at most 500 lines, headings, step lists, examples or code.
Hygiene15 / 15
No placeholders such as TODO, no Windows paths, with a licence.
Riskno signals detected in the scan
Files in this package
2
Scripts
0
HTML comments
0
URLs in SKILL.md
0
Deprecation

No signals detected. Repository changed on 26.10.2026, without references to retired models or APIs.

Licence

MIT, open: use and modify while retaining author attribution.

What it does and when to use it

Co-writing sessions under the No-Free-Zone protocol and feedback discipline, the Title Game (chained vowel-cascade title generation), title generation from an idea, and high-volume option dumps for ONE line or ONE section, with Pat Pattison's methods. Use when: 'co-write tonight', 'co-write rules', 'no-free-zone', 'Title Game', '10 titles fast', 'title candidates for X', 'give me 30 options for this line', 'brainstorm the whole chorus'. For blank-page/idea/fragment starts use /songwriting:workflow; for rhyme dumps use /songwriting:rhyme.

Triggers from the description

s methods. Use when:co-write rulesno-free-zoneTitle Game10 titles fasttitle candidates for Xgive me 30 options for this linebrainstorm the whole chorus

Arguments

/co-write [co-write|title-game|title|line-brainstorm|section-brainstorm] [args]
Original author description

Co-writing sessions under the No-Free-Zone protocol and feedback discipline, the Title Game (chained vowel-cascade title generation), title generation from an idea, and high-volume option dumps for ONE line or ONE section, with Pat Pattison's methods. Use when: 'co-write tonight', 'co-write rules', 'no-free-zone', 'Title Game', '10 titles fast', 'title candidates for X', 'give me 30 options for this line', 'brainstorm the whole chorus'. For blank-page/idea/fragment starts use /songwriting:workflow; for rhyme dumps use /songwriting:rhyme.

How it works

  1. Mandatory pre-flight. Response Filter
  2. Purpose
  3. Action Router
  4. Handlers
  5. Persistence and template overrides
  6. Boundary. This skill owns line emission, and pays for it
  7. Related skills
Original SKILL.md headings · 192 lines, 2,246 words
SKILL.md preview
---
description: "Co-writing sessions under the No-Free-Zone protocol and feedback discipline, the Title Game (chained vowel-cascade title generation), title generation from an idea, and high-volume option dumps for ONE line or ONE section, with Pat Pattison's methods. Use when: 'co-write tonight', 'co-write rules', 'no-free-zone', 'Title Game', '10 titles fast', 'title candidates for X', 'give me 30 options for this line', 'brainstorm the whole chorus'. For blank-page/idea/fragment starts use /songwriting:workflow; for rhyme dumps use /songwriting:rhyme."
argument-hint: "[co-write|title-game|title|line-brainstorm|section-brainstorm] [args]"
user-invocable: true
disable-model-invocation: false
---

## Mandatory pre-flight. Response Filter

Before emitting titles, line or section option dumps, or co-write feedback, run **§5 Title +
hook** and **§2 Line-writing** of
[response-filter](../../context/pat-pattison/research/response-filter.md) (add **§4 Coaching
posture** when facilitating a live session). NAME each box's pass / fail / skip-with-reason (aloud
or in reasoning); correct before emission. Skips are valid; silent skips are not.

§2's **Reference:** line is a load list, not a bibliography: §2 has not run until the files it
names were read this session, at minimum [meter](../../context/pat-pattison/research/meter.md)
and [phrasing](../../context/pat-pattison/research/phrasing.md) before the first candidate, plus
[metaphor](../../context/pat-pattison/research/metaphor.md) when the request asks for an image or
a figure, per [response-filter](../../context/pat-pattison/research/response-filter.md) "§2
Line-writing filter". Naming the §2 boxes without opening those files attests to the filter
instead of applying it, and that is what lets bad lines through: a miscounted stress, or a plain
description labeled a metaphor. Loading is an input, checked in the hard gate below.

When the output is a candidate LYRIC LINE about to be shown, cycle §2's boxes inside
[line-edit-rubric](../../context/pat-pattison/research/line-edit-rubric.md): load it and run the
FULL cycle per candidate, in its order, naming each pass. On fixed-melody work, show nothing
unless pass 1 (positional fit) comes back CLEAN.
*(Plugin-authored, not Pat Pattison's.)*

## Purpose

The high-volume options are raw material for the writer to choose from, not finished lines.

Method content is Pat Pattison's, under the plugin-root `../../context/pat-pattison/`; a future
author's method plugs in at `context/<author>/` without changing this skill, the author seam per
the plugin-root `../../README.md` "Method content and the author seam".

## Action Router

`/songwriting:co-write <action> [args]`. Parse `$ARGUMENTS`: first token = action when it matches
a listed action, remainder = args; otherwise treat all of `$ARGUMENTS` as payload for the default.
No action → open the co-write protocol (No-Free-Zone session opener).

| Action | Use when the user asks for | Load |
| --- | --- | --- |
| `co-write` (default) | co-writing session rules, feedback discipline, No-Free-Zone opener | [co-writing](../../context/pat-pattison/research/co-writing.md), [templates/co-write-session-opener](../../context/pat-pattison/templates/co-write-session-opener.md), [process](../../context/pat-pattison/research/process.md) |
| `title-game` | "Title Game", "Pat's title cascade", "10 titles fast", "vowel-chain titles", co-write warmup | [title-game](../../context/pat-pattison/research/title-game.md), [templates/title-game-prompt](../../context/pat-pattison/templates/title-game-prompt.md), [co-writing](../../context/pat-pattison/research/co-writing.md) |
| `title` | title candidates from an idea, title generation | [hook](../../context/pat-pattison/research/hook.md) "title generation", [idea-to-title](../../context/pat-pattison/research/idea-to-title.md), [templates/title-generation-prompt](../../context/pat-pattison/templates/title-generation-prompt.md) |
| `line-brainstorm` | "30 alternatives for this line", "more end-line words", "lots of swaps". HIGH VOLUME dump for ONE line | [line-brainstorm](../../context/pat-pattison/research/line-brainstorm.md), [templates/line-brainstorm-prompt](../../context/pat-pattison/templates/line-brainstorm-prompt.md), [mosaic-rhyme](../../context/pat-pattison/research/mosaic-rhyme.md), [meter](../../context/pat-pattison/research/meter.md), [phrasing](../../context/pat-pattison/research/phrasing.md), [line-edit-rubric](../../context/pat-pattison/research/line-edit-rubric.md) |
| `section-brainstorm` | "more options for the chorus", "high volume options for this section". Line-brainstorm per line + stability profile + hot-spot map | [line-brainstorm](../../context/pat-pattison/research/line-brainstorm.md) Scope B, [box-model](../../context/pat-pattison/research/box-model.md), [audit-checklist](../../context/pat-pattison/research/audit-checklist.md), [meter](../../context/pat-pattison/research/meter.md), [phrasing](../../context/pat-pattison/research/phrasing.md), [line-edit-rubric](../../context/pat-pattison/research/line-edit-rubric.md) |

## Handlers

- Facilitate a co-write as coaching, not monologue, under No-Free-Zone discipline: no idea is
  free, every suggestion earns its place; surface choice points, do not decide for the writers.
- **Line and section brainstorms: the DUMP goes to the file; the MENU goes to chat.** The 30-50+
  columns go to a labeled menu in `variations/` or `worksheets/`, never inline, and the display cap
  never lowers the generated volume. Chat gets 3-4 candidates, each a full section block in context
  with changed lines marked `►`, one labeled block per variation. A response with nothing singable
  in it has not been delivered. See
  [variations](../../context/pat-pattison/research/variations.md) "Presenting the candidates: chat
  vs file".
- **Rubric before the menu:** every candidate cycles all passes of
  [line-edit-rubric](../../context/pat-pattison/research/line-edit-rubric.md) before it is shown,
  and pass 1 must come back CLEAN, not merely run. A candidate that you flagged as failing a pass is
  **removed, not disclosed**: showing it with the flag attached is the rule violation, not the
  honest version of it. Present fewer candidates instead.
- **Two rejected executions in a slot stops generation.** *Slot* = the lyric position under
  revision: one line, or one section when the section is being rewritten whole; not the metrical
  or rhyme slot the craft files mean. *Rejection* = the writer declines the batch's EXECUTION:
  "none of these", "the idea's right, the lines aren't", or no candidate picked. Picking one and
  asking for a tweak is not a rejection, and rejecting the CONCEPT resets the count, because the
  next batch answers a different brief. On the second rejection do not produce a third batch; hand
  the concept over: (1) what the line has to do here, in one sentence; (2) the constraints, as the
  marked positional template plus the section's rhyme and sound obligations; (3) the unused raw
  material, naming images still sitting in the song's `ideation/`; (4) what the two rejected
  batches had in common, named as the failure mode; (5) one question for the writer. No candidate
  lines in that handoff. **Two** is the writer's threshold; do not raise it.

## Persistence and template overrides

Write generated files to the paths in
[artifact-persistence](../../context/pat-pattison/research/artifact-persistence.md) "Where
generated work persists", and honor a consuming project's own songwriting layout when it defines
one. Before loading any bundled `templates/<name>.md`, check
`${CLAUDE_PROJECT_DIR}/songwriting/templates/pat-pattison/<name>.md` first: a project-level
override wins over the bundled default, first match, per that file's "Template override".

## Boundary. This skill owns line emission, and pays for it

Every other craft skill routes finished lyric lines here, which makes this the one place generic
output escapes into a song. So the gate is an **input** gate: what must exist BEFORE lines are
written, not a checklist named afterward.

### Hard gate. Show the artifacts, do not claim them

A box is passed when its output is **visible in this response or on disk at a named path**. A box
named as passed with nothing to show is a failed box. For the source rows, what is shown is a
citation of what the file settled in THIS line; a filename is not an artifact, and neither is the
claim that the method is known. Every row applies before any line is emitted; the rhyme row
applies when the line sits in a rhyme position. The rows run in the order work happens: inputs,
then the rubric's recorded passes, then the skeptic's refutation from a fresh context.

| Required before emission | What "shown" means |
| --- | --- |
| Writer voiceprint loaded | the four dimensions visible in this response, or the path to the writer's voiceprint file. Built per [voiceprint](../../context/pat-pattison/research/voiceprint.md) |
| Craft sources read | `meter.md` and `phrasing.md` read this session, each cited by what it settled for this line, the stress count the line has to hit, the phrase shape it has to keep, not by filename |
| Metaphor source read, when a figure is asked for | `metaphor.md` read, and every figure emitted labeled with its type from Pat's three. Expressed Identity, Qualifying Metaphor, Verbal Metaphor. An unlabeled figure is a plain description until the label is shown |
| Sensory raw material exists | a path under the song's `ideation/`, or object-writing output in this response, from `/songwriting:object-writing generate` |
| Rhyme candidates generated | the labeled 8-15 candidate menu across ≥4 stability tiers, including ≥3 mosaic, visible, from `/songwriting:rhyme`, its search having walked the stressed vowel's coda field and not only the source word's own |
| Section mode named | stated in this response: does this section SHOW (verse) or TELL (chorus)? |
| Positional template or stress map marked | the marked line. Or, when a sung melody already exists, the numbered-and-bracketed positional template per [meter](../../context/pat-pattison/research/meter.md) "fitting a replacement line to an already-sung melody", never an assertion that it scans. From `/songwriting:meter-prosody` |
| Rubric cycled per candidate, pass 1 CLEAN | the pass-by-pass result visible in this response, or at the candidate's named path under the song's `variations/`, per [line-edit-rubric](../../context/pat-pattison/research/line-edit-rubric.md). Pass 1's artifact is the marked template above; on free-melody work the stress map stands in. Counts are run, not estimated: syllables per word from `${CLAUDE_PLUGIN_ROOT}/context/pat-pattison/scripts/datamuse.sh syllables <word>`, and pass 2's repeated words from a word-frequency count over the draft file (a `sort \| uniq -c` pipeline) |
| Skeptic refutation returned | the strongest case AGAINST each candidate, visible in this response or at a named path under the song's `variations/`, from a fresh subagent that READ [response-filter](../../context/pat-pattison/research/response-filter.md) §2, the marked map, and [line-edit-rubric](../../context/pat-pattison/research/line-edit-rubric.md) at those paths. A line survives when its refutation is stated and judged insufficient; a return holding nothing against anything has shown nothing |

**Every row except the rubric row may be skipped**, and a skip is **named, with its reason, in
the output**. A silent skip is the failure, and so is listing a box as passed while its artifact
does not exist. Those rows stay skippable because how much support a line gets is the writer's
craft call: "There are no rules, only tools." (*Writing Better Lyrics* (2009), Chapter 18, quoted
in [response-filter](../../context/pat-pattison/research/response-filter.md)). The skeptic row is
in that class deliberately: a refutation pass costs a subagent dispatch, and whether a batch is
worth one is a judgment, not a rule the writer laid down. Skip it by name and reason, and expect
to be asked why, because self-attestation is what failed.

**The rubric row is never skipped**: it is the writer's standing rule for every candidate, not a
craft box he may decline. Its record is not a verdict on the candidate: the countable passes run
as commands, and the skeptic row's fresh context tests the judgment passes. The rules behind it
are in [line-edit-rubric](../../context/pat-pattison/research/line-edit-rubric.md)
`## Standing operating rules`; in short, the cycle runs on every candidate every time with no
fatigue exception, and a FAILED pass kills the candidate rather than annotating it. Under load,
emit fewer candidates, not unchecked ones. Passes scoped out by their own headings (no fixed
melody, no rhyme position, no figure, no voiceprint on disk) are declared and stepped past: that is
the pass running, not a skip. The response filter's own §1-§8 boxes stay skippable-with-reason.

If a gate's input is missing, invoke the skill that produces it; do not proceed and note the
absence. Lines written without their inputs are LLM defaults wearing the method's vocabulary,
which is what the writer rejects. A source row's missing input is a file, so READ it: summarizing
what a craft file probably says, or working from a remembered version, is the same failure as an
unnamed skip.
*(Plugin-authored, not Pat Pattison's.)*

### Refute, do not review

*(Plugin-authored, not Pat Pattison's.)* A self-graded pass is indistinguishable from a real one
when the same context produced both: a model running the response filter on its own output passes
itself. So dispatch the skeptic to build the case AGAINST each candidate, and give it nothing but
the candidate batch, the slot's positional template or marked scansion map, and the source paths
to read: not the reasoning that produced the candidates, not which one you prefer. This is a hard
boundary, not a default, the same blind-dispatch mechanic that makes
`/songwriting:object-writing generate`'s fleet work. Its kill rules rank singability and verbosity
ABOVE cleverness: a clever line that cannot be sung in the slot's length is refuted, not merely
ranked lower. The pass runs before the writer sees the batch and settles nothing; the writer's
sing-check is still the last word. Any general subagent honors this today; no agent in this plugin
is a skeptic, and a preloaded-skill skeptic agent is deliberately not shipped here.

### Section mode binds before the line, not after

Verses show; the chorus tells. Verses carry specific situation, image, and action; the chorus makes
the broader statement the verses keep recoloring. Name the mode first, then check each emitted line
against it. A chorus of concrete inventory is verse material in the wrong box, however good the
inventory, and a chorus is sung back, so unsingable length is a defect, not a polish item.

### Mine → adapt → say it aloud

Object-writing output is ore. Pull ONE image forward and build the line around it; pasting a run of
sensory material into a section relocates the material, it does not use it. Running the first step
and skipping the other two is the failure this section prevents.

1. **Mine**. Take ONE image, verb, or detail off the page. A finished write is preparation, not
   draft: *"Once you become adept in your object writing, with bushels of sense-bound images
   glittering on the kitchen table, what do you do with them?"* (*Writing Better Lyrics* (2009),
   Chapter 2).
2. **Adapt**. Rewrite the mined image in plain sung English BEFORE it enters a slot. Quoting
   object-write texture into a lyric slot is what failed ("hung where it stops being anybody's" is
   ore that reads like a line). Pat's printed demonstration of the crafted shape is the collar:
   "Hot rod hearts and high school rings" replacing "All the things we used to do" (*Writing Better
   Lyrics* (2009), Chapter 2). That every word in it is a common one is this plugin's reading, not
   Pat's claim. **The step itself is plugin-authored:** Pat draws the raw-material-vs-crafted-line
   distinction and prints no translation procedure. See
   [object-writing](../../context/pat-pattison/research/object-writing.md).
3. **Say it aloud**. Would a person SAY the adapted line, talk-sung? Writer-derived from the same
   sessions; the box lives in
   [line-edit-rubric](../../context/pat-pattison/research/line-edit-rubric.md) pass 8. A line that
   fails it goes back to step 2, never forward to a rarer word.

## Related skills

- Sensory raw material before the lines → `/songwriting:object-writing generate`
- Metaphor for a line that needs one → `/songwriting:metaphor`
- Blank page, an idea/seed, or a stuck fragment → `/songwriting:workflow`
- Rhyme partners for the swaps → `/songwriting:rhyme`
- Pre-lock audit of the chosen line/section → `/songwriting:diagnose audit`
- The writer's own register, before any generation → build
  [voiceprint](../../context/pat-pattison/research/voiceprint.md)
- The per-candidate cycle every emitted line clears →
  [line-edit-rubric](../../context/pat-pattison/research/line-edit-rubric.md)
Files in this package (2)
  • SKILL.md 17 KB
  • evals/evals.json 3 KB

Versions

RepositoryScoreLicence
This page · melodic-software/claude-code-plugins84MIT
jkaloger/lazyspec100MIT

Database snapshot: 13.11.2026 · package version 9c7103705f38 (SHA-256 digest of the entire package)

Downloads in the last 30 days: 0 · ranking

Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.

Similar skills

  • 84song-form
    Song structure with Pat Pattison's methods: section identification, forms (AABA, verse/chorus, verse/refrain), the candy-bar rewrite, hook placement and hot spots, repetition and repainting (You-I-We, Past-Present-Future, hidden questions/commands), verse development, the box…
  • 100title-craft
    When the user wants to write YouTube video titles, optimize existing titles, A/B test title variants, or improve click-through rate. Also use when the user says 'write a title,' 'title ideas,' 'optimize my title,' 'fix this title,' 'title for my video,' 'improve CTR,' 'my titles…
  • 100yt-titles
    Generate high-performing YouTube title ideas based on the user's video content description, cross-referenced with your recent top-performing videos. Use this skill whenever the user wants YouTube title suggestions, asks "what should I title this video", "give me title ideas",…
  • 95title-tag
    When the user wants to write, audit, or optimize an HTML title tag, SEO title, page title, SERP title, browser-tab title, duplicate title, or title rewrite issue. For a sitewide metadata lifecycle including implementation and production validation, use page-metadata.
  • 100viral-title
    Generate high-potential viral title candidates for content across WeChat public account articles, X/Twitter posts, YouTube videos, Bilibili videos, and similar content platforms. Use when the user needs headline/title ideation, batch title generation, platform-specific title…
  • 100lw-rhyme
    Rhyme as a structural tool, not decoration (押韵与韵式). The full stability scale from perfect rhyme through family, additive, subtractive, assonance, consonance and partial rhyme - what each type does to closure - how to place stable and unstable types in the pushing and resolving…

Feedback and reports

Your message goes to Bartosz Sokoliński. An email address is needed only for a reply. See the privacy policy.