Writing staffing skill

Dispatch reference for composing writing teams.

by haowjy·Apache-2.0 license·★ 506 Stars on the repo·GitHub ↗

Use now

Files of Writing staffing

haowjy/main1 file shown
SKILL.md
Show the full text112 lines

Writing Staffing

Each subagent has its own skill set. This skill teaches what extra skills to load and reference when dispatching work.

Fan-out gives the same question and files to different model families for independent judgment. Reserve it for high-stakes calls where model diversity can reveal different blind spots. Parallel lanes use different prompts or focus areas; use them when the work divides cleanly.

Dispatch Reference

@writer

Extra skills: /character-sim for voice fidelity, /shared-dao for project vocabulary.

Reference: name the production mode from /creative-writing-modes → resources/prose-modes.md (fresh draft, revision, bridge, alternate take, line polish). Point to /creative-writing-craft → resources/prose-writing.md or resources/scene-construction.md when relevant. Include style and character state for everyone who appears, plus continuity anchors.

One writer per scene — voice consistency degrades when multiple writers handle adjacent content.

@critic

Extra skills: /creative-writing-craft for prose/voice focus, /shared-dao for vocabulary checks.

Assign a focus area: structure, character, voice, prose, or continuity. Include style for voice critique. Include character state for character or continuity focus.

Run different focus areas as parallel lanes. Scale to stakes: 1–2 for low-stakes, 3 for standard chapters, 4–5 for pivotal scenes with duplicated coverage on the critical dimension.

For a pivotal scene or disputed judgment, fan out the same critical dimension once across two strong model families, then synthesize the disagreement.

@editor

Name the edit level: editorial review, developmental, line edit, copyedit, proofreading. Point to /story-review → resources/editorial-review.md for holistic pass, or the specific edit-level resource.

Use when the draft needs a priority order across concerns. For depth on one dimension, use @critic.

@continuity-checker

Include the draft plus canon, timeline, character state, and vocab files. More expensive than a critic with continuity focus — reads broadly. Use the critic for routine checks, the continuity-checker for deep validation.

@brainstormer

Extra skills: /character-sim for character arcs, /creative-research for real-world grounding.

Run parallel lanes on different angles, not the same angle. Three perspectives beats five instances of one.

@outliner

Outlining starts after direction is chosen — use @brainstormer first. The outliner's output feeds the writer.

@style-creator

Include sample chapters or existing style files. Point to /creative-writing-craft → resources/style-analysis.md.

@reader-sim

Extra skills: /character-sim when the reader persona is a specific character type.

Specify the reader persona and knowledge boundary (what has this reader already read). Include the draft.

Run after the write/critique loop converges, before presenting to the author. A scene can be technically clean and leave a reader cold.

@character-sim

Include character state and style. Specify the scenario or relationship to explore. Use one parallel lane per character or perspective for independent exploration; use one shared simulation when testing their interaction.

@web-researcher

Specify the question, story context, and what the story currently assumes so the researcher can flag contradictions.

Effort Scaling

Scale critic coverage to stakes. Knowledge maintenance waits until direction or chapters settle. Reader-sim runs after the write/critique loop converges.

1---
2name: writing-staffing
3description: >
4 Dispatch reference for composing writing teams. Teaches which skills to
5 load for each subagent, which resources to reference, and when to fan out
6 versus run parallel lanes. Load when staffing a workflow.
7---
8 
9# Writing Staffing
10 
11Each subagent has its own skill set. This skill teaches what *extra* skills
12to load and reference when dispatching work.
13 
14**Fan-out** gives the same question and files to different model families for
15independent judgment. Reserve it for high-stakes calls where model diversity
16can reveal different blind spots. **Parallel lanes** use different prompts or
17focus areas; use them when the work divides cleanly.
18 
19## Dispatch Reference
20 
21### `@writer`
22 
23Extra skills: `/character-sim` for voice fidelity, `/shared-dao` for
24project vocabulary.
25 
26Reference: name the production mode from `/creative-writing-modes` →
27`resources/prose-modes.md` (fresh draft, revision, bridge, alternate take,
28line polish). Point to `/creative-writing-craft` →
29`resources/prose-writing.md` or `resources/scene-construction.md` when
30relevant. Include style and character state for everyone who
31appears, plus continuity anchors.
32 
33One writer per scene — voice consistency degrades when multiple writers
34handle adjacent content.
35 
36### `@critic`
37 
38Extra skills: `/creative-writing-craft` for prose/voice focus, `/shared-dao`
39for vocabulary checks.
40 
41Assign a focus area: structure, character, voice, prose, or continuity.
42Include style for voice critique. Include character state for
43character or continuity focus.
44 
45Run different focus areas as parallel lanes. Scale to stakes:
461–2 for low-stakes, 3 for standard chapters, 4–5 for pivotal scenes with
47duplicated coverage on the critical dimension.
48 
49For a pivotal scene or disputed judgment, fan out the same critical dimension
50once across two strong model families, then synthesize the disagreement.
51 
52### `@editor`
53 
54Name the edit level: editorial review, developmental, line edit, copyedit,
55proofreading. Point to `/story-review` → `resources/editorial-review.md` for
56holistic pass, or the specific edit-level resource.
57 
58Use when the draft needs a priority order across concerns. For depth on
59one dimension, use `@critic`.
60 
61### `@continuity-checker`
62 
63Include the draft plus canon, timeline, character state, and vocab files.
64More expensive than a critic with continuity focus — reads broadly. Use the
65critic for routine checks, the continuity-checker for deep validation.
66 
67### `@brainstormer`
68 
69Extra skills: `/character-sim` for character arcs, `/creative-research` for
70real-world grounding.
71 
72Run parallel lanes on different *angles*, not the same angle. Three perspectives
73beats five instances of one.
74 
75### `@outliner`
76 
77Outlining starts after direction is chosen — use `@brainstormer` first.
78The outliner's output feeds the writer.
79 
80### `@style-creator`
81 
82Include sample chapters or existing style files. Point to
83`/creative-writing-craft` → `resources/style-analysis.md`.
84 
85### `@reader-sim`
86 
87Extra skills: `/character-sim` when the reader persona is a specific
88character type.
89 
90Specify the reader persona and knowledge boundary (what has this reader
91already read). Include the draft.
92 
93Run after the write/critique loop converges, before presenting to the
94author. A scene can be technically clean and leave a reader cold.
95 
96### `@character-sim`
97 
98Include character state and style. Specify the scenario or
99relationship to explore. Use one parallel lane per character or perspective
100for independent exploration; use one shared simulation when testing their
101interaction.
102 
103### `@web-researcher`
104 
105Specify the question, story context, and what the story currently assumes
106so the researcher can flag contradictions.
107 
108## Effort Scaling
109 
110Scale critic coverage to stakes. Knowledge maintenance waits until direction
111or chapters settle. Reader-sim runs after the write/critique loop converges.
112 

Discussion