← Back to Skills Marketplace
nitishgargiitd

Docs Cog

by CellCog · GitHub ↗ · v1.0.13 · MIT-0
darwinlinuxwindows ✓ Security Clean
3268
Downloads
6
Stars
18
Active Installs
14
Versions
Install in OpenClaw
/install docs-cog
Description
AI document generation powered by CellCog — PDF and DOCX. Resumes, contracts, reports, proposals, invoices, certificates, NDAs, letters, brochures, legal doc...
README (SKILL.md)

Docs Cog - Professional Documents Powered by CellCog

Deep reasoning. Accurate data. Beautiful design. The three things every great document needs — and most AI gets wrong.

Professional document generation — PDF by default, native DOCX on request. Deep research fills content from minimal prompts. Resumes, contracts, reports, proposals — delivered in minutes.

How to Use

For your first CellCog task in a session, read the cellcog skill for the full SDK reference — file handling, chat modes, timeouts, and more.

OpenClaw (fire-and-forget):

result = client.create_chat(
    prompt="[your task prompt]",
    notify_session_key="agent:main:main",
    task_label="my-task",
    chat_mode="agent",
)

All agents except OpenClaw (blocks until done):

from cellcog import CellCogClient
client = CellCogClient(agent_provider="openclaw|cursor|claude-code|codex|...")
result = client.create_chat(
    prompt="[your task prompt]",
    task_label="my-task",
    chat_mode="agent",
)
print(result["message"])

Output Formats

CellCog generates professional documents in multiple formats:

PDF (Default for Ambiguous Requests)

When you say "create a report" or "make a document" without specifying a format, CellCog generates PDF — perfect typography, layouts, and design with full creative control.

DOCX (First-Class Support)

When you explicitly request DOCX ("create a Word document", "make a .docx"), CellCog generates native Word directly — editable, compatible with Microsoft Word and Google Docs, great for collaborative workflows where multiple people contribute.

Just ask for DOCX and you'll get it. No need to justify why you want an editable format.


What Documents You Can Create

Resume & Career Documents

Build your professional story:

  • Resume/CV: "Create a modern resume for a software engineer with 5 years of experience"
  • Cover Letter: "Write a compelling cover letter for a product manager position at Google"
  • LinkedIn Summary: "Create a professional LinkedIn summary that highlights my transition from finance to tech"
  • Portfolio: "Build a portfolio document showcasing my UX design projects"

Example prompt:

"Create a modern, ATS-friendly resume for:

Name: Sarah Chen Title: Senior Product Manager Experience: 7 years in B2B SaaS

Work history:

  • Stripe (2021-present): Led payments platform, grew revenue 40%
  • Slack (2018-2021): Launched enterprise features
  • Microsoft (2016-2018): Associate PM on Azure

Education: Stanford MBA, UC Berkeley CS

Clean, professional design with blue accents."

Business Documents

Professional business materials:

  • Proposals: "Create a consulting proposal for a digital transformation project"
  • Invoices: "Generate an invoice template for my freelance design business"
  • Business Letters: "Write a formal partnership inquiry letter"
  • Quotes & Estimates: "Create a detailed quote for website development services"
  • Meeting Minutes: "Format these meeting notes into professional minutes"

Example prompt:

"Create a business proposal for 'CloudMigrate' consulting services:

Client: Acme Corp Project: AWS cloud migration Timeline: 6 months Budget: $150,000

Include: Executive summary, scope of work, timeline, team bios, pricing breakdown, terms.

Professional, corporate design."

Reports & Analysis

Data-driven documents:

  • Business Reports: "Create a quarterly business review report"
  • Research Reports: "Format my research findings into a professional report"
  • Analysis Documents: "Create a competitive analysis document"
  • White Papers: "Build a white paper on AI in healthcare"
  • Case Studies: "Create a customer case study showcasing ROI"

Example prompt:

"Create a Q4 2025 business report:

Title: Quarterly Performance Review Company: TechStart Inc.

Key metrics:

  • Revenue: $2.1M (up 35% YoY)
  • Customers: 150 (up from 98)
  • Churn: 5% (down from 8%)

Include charts and executive summary. Corporate professional style."

Legal & Finance Documents

Formal agreements and contracts:

  • Contracts: "Create a freelance services agreement"
  • NDAs: "Generate a mutual non-disclosure agreement"
  • Terms of Service: "Draft terms of service for my SaaS app"
  • Privacy Policies: "Create a GDPR-compliant privacy policy"
  • MOUs: "Create a memorandum of understanding between two companies"

Example prompt:

"Create a freelance contractor agreement:

Client: Acme Corp Contractor: Jane Smith (Web Developer) Project: E-commerce website redesign Duration: 3 months Payment: $15,000 (50% upfront, 50% on completion)

Include: Scope, deliverables, payment terms, IP ownership, confidentiality, termination clauses.

Professional legal formatting."

Creative & Marketing Documents

Eye-catching marketing materials:

  • Brochures: "Create a tri-fold brochure for our fitness studio"
  • Flyers: "Design a promotional flyer for our summer sale"
  • One-Pagers: "Create a product one-pager for sales team"
  • Media Kits: "Build a media kit for our startup"
  • Catalogs: "Create a product catalog with 20 items"

Example prompt:

"Create a product one-pager for 'TaskFlow' project management software:

Headline: Finally, a PM tool that doesn't suck Key features: AI task prioritization, Slack integration, real-time collaboration Pricing: $12/user/month Call to action: Start free trial

Modern, bold design. Blue and white color scheme."

Education & Training Documents

Learning materials:

  • Lesson Plans: "Create a lesson plan for teaching Python basics"
  • Training Manuals: "Build an employee onboarding manual"
  • Worksheets: "Create practice worksheets for algebra"
  • Course Outlines: "Design a 12-week data science curriculum"
  • Study Guides: "Create a study guide for AWS certification"

Events & Planning Documents

Event materials:

  • Invitations: "Create elegant wedding invitations"
  • Event Programs: "Design a conference program booklet"
  • Agendas: "Create a workshop agenda document"
  • Itineraries: "Build a detailed travel itinerary"
  • Certificates: "Create achievement certificates for our hackathon"

Forms & Certificates

Official documents:

  • Certificates: "Create a course completion certificate"
  • Awards: "Design employee of the month award"
  • Badges: "Create digital badges for our training program"
  • Forms: "Design a customer feedback form"

Chat Mode for Documents

Scenario Recommended Mode
Standard documents - resumes, invoices, reports, certificates "agent"
Complex documents requiring narrative craft - proposals, white papers, case studies "agent team"

Use "agent" for most documents. Resumes, contracts, reports, and standard business documents execute well in agent mode.

Use "agent team" for high-stakes documents where persuasion and narrative flow matter—investor proposals, detailed white papers, compelling case studies.


Tips for Better Documents

  1. Provide the content: Don't say "write about my experience" - provide actual details, numbers, and facts.

  2. Specify structure: "Include: Executive summary, problem, solution, timeline, pricing" gives clear direction.

  3. Design preferences: "Modern and minimal", "Corporate professional", "Bold and colorful" - describe what you want.

  4. Brand elements: Mention colors, logos (upload them), or reference existing brand guidelines.

  5. Audience context: "For investors", "For enterprise clients", "For students" changes tone and detail level.

  6. Choose your format: PDF is the default for polished output. Request DOCX when your team needs to edit or customize the document.


If CellCog is not installed

Run /cellcog-setup (or /cellcog:cellcog-setup depending on your tool) to install and authenticate. OpenClaw users: Run clawhub install cellcog instead. Manual setup: pip install -U cellcog and set CELLCOG_API_KEY. See the cellcog skill for SDK reference.

Usage Guidance
This skill appears coherent for its stated purpose, but before installing you should: 1) Verify CELLCOG_API_KEY is obtained from the official CellCog service and scope/permissions are understood; 2) Confirm you trust https://cellcog.ai (review privacy/terms) because document content (which may include sensitive data) will be sent to that external service; 3) Ensure python3 and the 'cellcog' Python package (or equivalent runtime SDK) are installed from a trusted source (PyPI/GitHub official repo) since the skill provides no install steps; 4) Avoid submitting highly sensitive PII or secrets in prompts unless you’re comfortable with CellCog’s handling of user data; 5) If you want to prevent autonomous calls to the external API, restrict the agent’s ability to invoke skills autonomously at the platform level.
Capability Analysis
Type: OpenClaw Skill Name: docs-cog Version: 1.0.13 The docs-cog skill is a legitimate integration for the CellCog document generation service, allowing users to create professional PDFs and DOCX files. The SKILL.md file provides standard usage instructions, example prompts for various document types (resumes, contracts, reports), and installation steps for the required 'cellcog' Python dependency. No evidence of data exfiltration, malicious execution, or harmful prompt injection was found.
Capability Tags
cryptocan-make-purchasesrequires-sensitive-credentials
Capability Assessment
Purpose & Capability
The name/description (document generation) aligns with the declared requirements: python3 and a CELLCOG_API_KEY are reasonable for a Python-based SDK that calls an external CellCog service. Minor inconsistency: metadata lists a dependency 'cellcog' but there is no install spec — this is plausible for an instruction-only skill (it expects the runtime to already have the SDK), but the skill does not provide an installation mechanism for that dependency.
Instruction Scope
SKILL.md contains only usage examples for the CellCog SDK (client.create_chat) and guidance about formats and prompts. It does not instruct the agent to read unrelated files, environment variables, or system paths, nor to transmit data to unexpected endpoints beyond the CellCog API. It does reference a separate 'cellcog' skill for fuller SDK docs, which is consistent with a multi-skill ecosystem.
Install Mechanism
No install specification is present (instruction-only), which is low risk. Because there is no download/extract/install step, nothing will be written to disk by the skill itself. The only potential friction is that the SKILL.md assumes a 'cellcog' Python package is available but does not provide how to install it; that's an operational omission, not a security red flag.
Credentials
Only one required environment variable (CELLCOG_API_KEY) is declared, which is appropriate for an API-backed document generation service. No unrelated secrets or system credentials are requested.
Persistence & Privilege
The skill does not request persistent/always-on inclusion (always: false). It does not ask to modify other skills' configs or claim elevated agent privileges. Autonomous invocation (disable-model-invocation: false) is the platform default and not by itself a concern here.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install docs-cog
  3. After installation, invoke the skill by name or use /docs-cog
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.13
- Added environment requirements for skill operation: now specifies that `python3` must be available and the `CELLCOG_API_KEY` environment variable must be set. - No changes to usage instructions, feature set, or output formats.
v1.0.12
- Shortened and simplified the description for clarity and conciseness. - Updated agent usage instructions for clarity: now specifies "All agents except OpenClaw" instead of naming specific agents. - Added "legal documents" and "cover letters" to the list of supported document types in the description. - Improved prompt examples and formatting tips for easier understanding. - No functional or code changes—documentation update only.
v1.0.11
- Updated description to emphasize AI generation, expanded supported formats, and recent benchmark achievement (#1 on DeepResearch Bench, Apr 2026). - Added full Python usage example for integration with various agent providers. - Clarified output formats and simplicity of requesting DOCX files. - No code or interface changes—documentation improvement only.
v1.0.10
- Documentation was significantly condensed for clarity and brevity. - The description now summarizes features in a more concise form. - Prerequisites and usage instructions are simplified and streamlined. - Unnecessary repetition and long-form marketing claims were removed. - Example prompts and document type lists are retained but tightened. - Tips for better document creation are presented in a more direct format.
v1.0.9
Major SKILL.md rewrite — improved clarity, examples, and coverage. - SKILL.md completely rewritten for clarity, depth, and real-world prompts. - Expanded examples with sample prompts for resumes, contracts, reports, proposals, and more. - Clarified output format logic (PDF by default, DOCX on request). - Added new sections: Chat mode recommendations and detailed document categories. - Updated description to reflect CellCog’s benchmark ranking and document quality. - Old content reorganized and expanded with practical usage advice.
v1.0.8
- SKILL.md updated with a shorter, more focused overview and new supported formats (PDF, DOCX, PPTX, XLSX). - Expanded description of document types and added tables showing output format guidance. - Lists CellCog’s internal capabilities, including PDF/Word/PowerPoint/Excel generation, deep research, and image support. - Adds a “Related Skills” section with links for specialized document needs. - Example prompts, design explanations, and detailed document guides were removed for brevity.
v1.0.7
- Clarified and updated the Python example code for skill setup, distinguishing between OpenClaw agents (with `notify_session_key`) and other agents. - Improved prerequisite section for clarity on agent modes and blocking/non-blocking behavior. - Updated references to the cellcog skill for SDK/API documentation. - No functional or API changes—documentation updates only.
v1.0.6
docs-cog 1.0.6 Changelog - Updated SKILL.md for improved clarity and conciseness. - Rewrote the "Quick pattern" in the prerequisites to a simpler "Quick start" code example. - Added notes directing users to the cellcog skill for full SDK API details and delivery modes. - No changes to document capabilities, APIs, or output formats; documentation streamlined only.
v1.0.5
- Updated DeepResearch Bench date reference from "Feb 2026" to "Apr 2026" in both the description and documentation. - No other changes; documentation content remains the same.
v1.0.4
- Updated the SKILL.md description for clarity and brevity. - Added metadata fields: supported OS list and homepage URL. - No changes to core functionality; documentation improvement only.
v1.0.3
- Adds explicit, first-class support for native DOCX (Word) output in addition to PDF. - Updated documentation to clarify output format selection — PDF remains default, DOCX available upon request. - Enhanced descriptions to reflect seamless generation of editable Word documents for collaborative workflows. - No code changes; documentation improvements only.
v1.0.2
- Added author and dependencies fields to skill metadata for improved documentation clarity. - Updated prerequisites section to refer to the skill as `cellcog` for consistency. - No changes to functionality or document creation workflows.
v1.0.1
- Updated the description and introduction to emphasize deep reasoning, factual accuracy, and advanced design; highlighted #1 ranking on DeepResearch Bench (Feb 2026) and state-of-the-art PDF generation. - Improved marketing language sitewide to stress CellCog’s document quality, accuracy, and SOTA (state-of-the-art) search models. - No changes to core usage instructions, supported document types, or technical prerequisites. - Documentation now positions the skill more competitively for professional and enterprise audiences.
v1.0.0
Initial release: AI-powered document generation using CellCog mothership skill. - Instantly create polished PDF documents: resumes, contracts, reports, marketing materials, legal docs, business proposals, and more. - Supports professional, customizable templates and layouts—specify structure, content, and design preferences. - Built-in prompt patterns for quick integration with seamless PDF generation (DOCX on request). - Simple mode guidance: use "agent" for most tasks; "agent team" for complex, narrative-driven documents. - Requires CellCog skill for SDK/API access.
Metadata
Slug docs-cog
Version 1.0.13
License MIT-0
All-time Installs 18
Active Installs 18
Total Versions 14
Frequently Asked Questions

What is Docs Cog?

AI document generation powered by CellCog — PDF and DOCX. Resumes, contracts, reports, proposals, invoices, certificates, NDAs, letters, brochures, legal doc... It is an AI Agent Skill for Claude Code / OpenClaw, with 3268 downloads so far.

How do I install Docs Cog?

Run "/install docs-cog" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Docs Cog free?

Yes, Docs Cog is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Docs Cog support?

Docs Cog is cross-platform and runs anywhere OpenClaw / Claude Code is available (darwin, linux, windows).

Who created Docs Cog?

It is built and maintained by CellCog (@nitishgargiitd); the current version is v1.0.13.

💬 Comments