prompt-manager
Agent for managing AI prompts on prompts.chat - search, save, improve, and organize your prompt library.
Flowy lists 232 subagents for Content, and the most-starred are prompt-manager (from the 174,766-star promptschat plugin), skill-manager (from the 174,766-star promptschat plugin) and humanize-diagnostician (from the 6,253-star humanize-korean plugin). The newest push among those Content subagents, on October 27, 2026, went to the promptschat plugin.
Agent for managing AI prompts on prompts.chat - search, save, improve, and organize your prompt library.
Agent for managing AI Agent Skills on prompts.chat - search, create, and manage multi-file skills for Claude Code.
์ ๋ฐ(strict) ๋ชจ๋ 1๋จ๊ณ ์ง๋จ ์์ด์ ํธ. ๊ธ ์ ์ฒด๋ฅผ ํ ๋ฒ์ ๋ณด๊ณ "๊ฐ์ฅ ์ง๋ฐฐ์ ์ธ AI ํฐ ํจํด 3~6๊ฐ"๋ฅผ taxonomy ID์ ํจ๊ป ์ง๋จํ๋ค. ๋ถ์์ ํ span ์ด๊ฑฐ(0โ18๊ฐ๋ก ์๋) ๋์ "๋ฌด์์ด ์ด ๊ธ์โฆ
์ ๋ฐ(strict) ๋ชจ๋ 3๋จ๊ณ ๋ง๋ฌด๋ฆฌ ์์ด์ ํธ. ์๋ฌธ๊ณผ ์ค๋ฌธ๋ณธ์ ์ง์ ๋์กฐํด โ ์๋ฏธ ๋ณด์กด(15ํญ โ ๊ฐ์ฃผยท์ ๋ชฉยท์๋ ์ฃผ์ฅ ์ฃผ์ ํฌํจ) โก์์ฐ์ฑ(์์กด AI ํฐ + ๊ณผ์ค๋ฌธ ์๋ฐฉํฅ)์ ํ ์ฝ๋ก ๋ณํฉ ํ์ ํ๊ณ , ๋ฌธ์ ๊ตฌ๊ฐ๋ง ๊ตญ์โฆ
์ ๊ฒฝ๋ก(lightยทstandardยทheavy) ๊ณต์ฉ ๋จ์ผ ํธ์ถ ์ค๋ฌธ ์์ด์ ํธ. ํ ํธ์ถ ์์์ ํ์งยท์ค๋ฌธยท์์ฒด๊ฒ์ฆ์ ์ผ๊ด ์ํํ์ฌ 5,000์ ์ดํ ํ๊ธ ์ ๋ ฅ์ 2~3๋ถ ์์ ์ฒ๋ฆฌํ๋ค. ์ฐ์ถ๋ฌผ์ final.md 1๊ฐ(๋ณธ๋ฌธ ๋์โฆ
Topic Clusters curator for the Claude Blog Brain. Maintains and answers from the Topic Clusters theme of the brain, grounded inโฆ
Data Integrations curator for the Claude Blog Brain. Maintains and answers from the Data Integrations theme of the brain,โฆ
Distribution curator for the Claude Blog Brain. Maintains and answers from the Distribution theme of the brain, grounded in theโฆ
Optional gallery research helper. Use when a caller wants to find references, explore styles, build a mood board, or needsโฆ
Optional image execution helper for callers that choose to delegate. Accepts a complete generation request and returns its resultโฆ
Optional prompt writing helper. Use only when the caller requests multiple distinct prompts at once โ for parallel imageโฆ
Weekly movement briefing on the AI agent harness space โ what climbed, what entered the radar, what died โ from theโฆ
Picks the right AI agent harness for a described task or project. Use when the user asks "what harness/framework/agent toolโฆ
Audits a codebase's AI agent stack against the live best-of-Agent-Harnesses dataset โ finds which harnesses the repo uses, flagsโฆ
Freeze a visual brief and compile bounded, model-aware prompts for complex, branded, text-heavy, or ambiguous image work. Useโฆ
Independently inspect generated or edited image files against a frozen brief. Use after generation for high-value, branded,โฆ
ๆ ็ถๆๆง่กๅๅ ๏ผๅฎๆๅไธไปปๅกๅ่ฟๅ็ปๆๅ็ปๆ
ๆธฒๆๅๆ็คบ่ฏ๏ผๅๆบ้ชๆถๅบ็บฟ๏ผ่ฏปๅ็ซ vol-{N}-ch-{M}-prompt.md๏ผ
ไฝ่ ๆฒ้็จ่ฏญ่ง่๏ผๅฏนไฝ่ ๅฑ็คบ็จๅคง็ฝ่ฏ + ๆฏ่ฏญๅฏน็ ง่กจ๏ผ
Expert code reviewer who provides constructive, actionable feedback focused on correctness, maintainability, security, andโฆ
ๆธฒๆๅๆ็คบ่ฏ๏ผๅๆบ้ชๆถๅบ็บฟ๏ผ่ฏปๅ็ซ vol-{N}-ch-{M}-prompt.md๏ผ
ไฝ่ ๆฒ้็จ่ฏญ่ง่๏ผๅฏนไฝ่ ๅฑ็คบ็จๅคง็ฝ่ฏ + ๆฏ่ฏญๅฏน็ ง่กจ๏ผ
Expert code reviewer who provides constructive, actionable feedback focused on correctness, maintainability, security, andโฆ
The dependency's `handleToNative.setProperties` reads six descriptor fields with `ctx.getProp`. Its early returns for `undefined`โฆ
ๆ็ซ ่ง่ง็ญๅไธ้ ๅพๅญไปฃ็ใ็ฑๅทฅไฝๆตๅฏผๆผๅจ Humanizer ไนๅใๆ็ปไบๅฎๆ ธๆฅไนๅ็ Stage 11 ๆพๅผ่ฐ็จใ
[Subagent] ๅไฝๅค็ๅญฆไน ๅจใ ๅฏนๆฏ AI ๅ็จฟ(draft_v1.md) ไธ็จๆท็กฎ่ฎค็ๆ็ปๅฎ็จฟ๏ผๆ็ผ็ปๆๅ็ๅไฝ็ป้ชๆ่ฎญใ ็ฑๅทฅไฝๆตๅฏผๆผๅจๆต็จๆถๅฐพ้ถๆฎตๅง็ป่ฐ็จ๏ผๆฒกๆๅฏๅญฆไน ๅทฎๅผๆถไน่ฝ็่ฎฐๅฝๅๅ ใ
Validates and structures lesson candidates before persisting to lessons.json
`code-surgeon-agent` ๆฏ PaperFit ็ๆบ็ ๆง่กๅจใๅฎ่ด่ดฃๆไธๆธธๅทฒ็ปๅณๅฎ็ไฟฎๅค็ญ็ฅๅฎๅ จๅฐ่ฝๅฐ `.tex` ไธ๏ผๅๆไฟฎๆน็ปๆไบคๅ้ญ็ฏใ
`layout-detective-agent` ๆฏ PaperFit ็่ง่ง่ฏๆญ Agentใๅฎ็่่ดฃๅชๆไธไธช๏ผๅบไบ PDF ้กตๅพใไบคๅๅผ็จไธๆฅๅฟ็บฟ็ดข๏ผๅคๆญ่ฎบๆๅฝๅๅฐๅบๆๅชไบๅฏ่งๆ็้ฎ้ข๏ผไปฅๅ่ฟไบ้ฎ้ข็ไธฅ้็จๅบฆใ
ไฝ ๆฏ **Orchestrator Agent**๏ผไธป่ฐๅบฆๅจ๏ผ๏ผๆฏ PaperFit ็ณป็ป็ไธญๅคฎๅ่ฐ่ ใไฝ ็ๆ ธๅฟ่่ดฃๆฏ๏ผ
showtime crew. Dispatched by the showtime skill, not for general requests. Use when a video job needs a brand kit extracted fromโฆ
showtime crew. Dispatched by the showtime skill, not for general requests. Use when a video job needs a brand kit extracted fromโฆ
showtime crew. Dispatched by the showtime skill, not for general requests. Use in studio or publish-bound video jobs with an openโฆ
Content strategist for canvas text nodes. Writes titles, descriptions, annotations, labels, and structured content for all canvasโฆ
Spatial positioning specialist for Obsidian Canvas. Computes optimal node positions using layout algorithms (dagre, grid, radial,โฆ
Media asset dispatcher for canvas generation. Generates images via /banana (Gemini), SVG diagrams via /svg, and GIFs via gifโฆ
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
By the GitHub stars of the plugin each ships in (Flowy does not rate quality), the best Content subagent for Claude Code is prompt-manager (from the 174,766-star promptschat plugin), level on stars with skill-manager (from the 174,766-star promptschat plugin), ahead of humanize-diagnostician (from the 6,253-star humanize-korean plugin).
Flowy lists 232 subagents for Content, and this page draws the top 47 Content subagents and leaves out 1 documentation file the index read as subagents (MUTATION_TESTING).