Manus

What it does

When your agent uses it

• User says use manus
• Delegate to manus
• Check manus sessions
• Tasks require deep web research

Example prompts

• “use manus”
• “delegate to manus”
• “send to manus”
• “/manus”

Requirements

• A credential in MANUS_API_KEY
• A credential in API_KEY

Workflow steps

3 steps, taken from the first numbered list in SKILL.md.

1. 1Create task → Returns task_id immediately
2. 2Poll status → Check until status is completed or failed
3. 3Return results → Extract output text and file attachments

What it can do on your machine

Read from SKILL.md and the folder at commit 281d88d. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Shell commands in SKILL.md call:

  • curl
  • jq

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  Hosts in commands or code, which the agent is likely to contact:

  • api.manus.ai
  • manus.im

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names these keys or tokens, usually read from environment variables:

  • MANUS_API_KEY
  • API_KEY

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Manus loads about 1.7k tokens when it runs, and up to ~2.3k if it reads all its reference files. Until then it costs about 115 tokens; SKILL.md has 396 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~115

When it runs · the whole SKILL.md, loaded when a task matches

~1.7k

With references · SKILL.md plus every file in references/, read only if the agent opens them

~2.3k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from sanjay3290/ai-skills at commit 281d88d, republished under its Apache-2.0 licence (© sanjay3290). 396 words, ~1,698 tokens.

name

manus

description

Delegate complex, long-running tasks to Manus AI agent for autonomous execution. Use when user says 'use manus', 'delegate to manus', 'send to manus', 'have manus do', 'ask manus', 'check manus sessions', or when tasks require deep web research, market analysis, product comparisons, stock analysis, competitive research, document generation, data analysis, or multi-step workflows that benefit from autonomous agent execution with parallel processing.

license

Apache-2.0

metadata.author

sanjay3290

metadata.version

1.0

Find the best 4K monitors for Mac with Thunderbolt connectivity and 120Hz+ refresh rate.
Focus on BenQ, Samsung, Dell, LG. Only include models released in the last year.
Compare specs, prices, and Mac-specific features. Output a comparison table.

Analyze [TICKER] stock: company profile, recent performance, technical indicators,
valuation metrics, and insider activity. Include a price chart for the past year.

Research the [industry] market: key players, market size, growth trends,
recent developments, and competitive landscape. Focus on [region/segment].

Compare [Product A] vs [Product B] vs [Product C]: features, pricing,
user reviews, pros/cons. Create a decision matrix for [use case].

curl -s -X POST "https://api.manus.ai/v1/tasks" -H "API_KEY:$MANUS_API_KEY" -H "Content-Type:application/json" -d '{"prompt":"<task description>","agentProfile":"manus-1.6"}'

{"task_id":"abc123","task_title":"...","task_url":"https://manus.im/app/abc123"}

curl -s -X GET "https://api.manus.ai/v1/tasks/{task_id}" -H "API_KEY:$MANUS_API_KEY"

curl -s -X GET "https://api.manus.ai/v1/tasks/{task_id}" -H "API_KEY:$MANUS_API_KEY" | jq -r '.output[] | select(.role=="assistant") | .content[] | select(.type=="output_text") | .text'

curl -s -X GET "https://api.manus.ai/v1/tasks/{task_id}" -H "API_KEY:$MANUS_API_KEY" | jq -r '.output[] | select(.role=="assistant") | .content[] | select(.type=="output_file") | "\(.fileName): \(.fileUrl)"'

curl -s -X GET "https://api.manus.ai/v1/tasks" -H "API_KEY:$MANUS_API_KEY" | jq '.data[] | {id, status, title: .metadata.task_title}'

curl -s -X POST "https://api.manus.ai/v1/tasks" -H "API_KEY:$MANUS_API_KEY" -H "Content-Type:application/json" -d '{"prompt":"follow-up question","taskId":"abc123","agentProfile":"manus-1.6"}'

curl -s -X DELETE "https://api.manus.ai/v1/tasks/{task_id}" -H "API_KEY:$MANUS_API_KEY"

{"prompt":"Analyze this","attachments":[{"type":"url","url":"https://example.com/doc.pdf"}]}

{"prompt":"What's in this image?","attachments":[{"type":"base64","data":"<base64>","mime_type":"image/png"}]}

{"prompt":"Review this file","attachments":[{"type":"file","file_id":"file-xxx"}]}

curl -s -X POST "https://api.manus.ai/v1/projects" -H "API_KEY:$MANUS_API_KEY" -H "Content-Type:application/json" -d '{"name":"My Project","instruction":"Always respond concisely"}'

curl -s -X POST "https://api.manus.ai/v1/tasks" -H "API_KEY:$MANUS_API_KEY" -H "Content-Type:application/json" -d '{"prompt":"...","projectId":"proj_xxx","agentProfile":"manus-1.6"}'

curl -s -X GET "https://api.manus.ai/v1/tasks/{task_id}" -H "API_KEY:$MANUS_API_KEY" | jq '{status, error}'

© sanjay3290, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file (references) in skills/manus of sanjay3290/ai-skills.

• SKILL.md
• references/api.md

Open the folder on GitHubat commit 281d88d