Skip to main content

Autonomous AI SEO + Sales Agent System built on OpenClaw

Project description

KIESSCLAW

CLI Prompt + Code Generator for Agentic Workflows Built on OpenClaw principles · Deterministic scaffolds · OSS-first

PyPI CI Docs License: MIT Python 3.10+ OpenClaw Compatible


What is KIESSCLAW?

KIESSCLAW is an all-in-one CLI for generating agent/pipeline/workflow scaffolds and prompt packs.
It ships with opinionated use cases and strict templates so teams can go from idea to runnable project structure in minutes.

Built by Erwin Kiess-Alfonso as part of the KLYTICS open-source ecosystem.


The Agent Team

Agent Codename Role Model
KiessProspector KPRO Lead import, enrichment, ICP scoring gpt-4o
KiessSequencer KSEQ Sequence orchestration, scheduling, safety gpt-4o-mini
KiessPen KPEN Personalization, first lines, templates gpt-4o
KiessInbox KINB Reply classification, qualification, DNC gpt-4o
KiessMetrics KMET Analytics, deliverability, reporting gpt-4o-mini
KiessCrawler KRAWL Technical SEO crawl, score, and issues gpt-4o
KiessAnalyst KANA Keyword + SERP + competitor gap analysis gpt-4o
KiessWriter KWRI SEO briefs + metadata generation gpt-4o-mini
KiessScout KSCO Backlink prospecting + link gap targeting gpt-4o-mini
KiessReporter KREP Weekly SEO reporting gpt-4o-mini

All agents share one workspace brain — a markdown-based memory system where context persists, logs are written, and data is stored.


Features

  • Use Case Registry — 6 built-in workflows (outreach_sdr, seo_audit_to_leads, github_oss_launch, etc.)
  • Prompt Factory — Render deterministic system, task, guardrails, and checklist prompts
  • Prompt Linting — Detect unresolved placeholders, vague instructions, and missing success criteria
  • Code Scaffolding — Generate agent/skill/workflow/test/docs skeletons safely
  • No-Overwrite Safety — Files are never replaced unless --force is explicitly passed
  • Doctor Command — Config/workspace/dependency diagnostics with quick fixes
  • Existing SDR + SEO Engine — Operational agents and API still available for runtime workflows

Quick Start

Prerequisites

  • Python 3.10+
  • An OpenAI API key (or Anthropic/Ollama)

Install

pip install kiessclaw
pip install kiessclaw[all]   # includes anthropic + slack
# Or install from source
git clone https://github.com/monykiss/kiessclaw.git
cd kiessclaw
pip install -e .

# Copy and configure environment
cp .env.example .env
nano .env  # Add your API keys

# Initialize workspace
python scripts/init_workspace.py

# Check status
python -m kiessclaw status

Docker

docker-compose up
# API live at http://localhost:8000
# Docs at http://localhost:8000/docs

Quick Start (Generator Wedge)

# 1) List built-in use cases
kiessclaw usecase list

# 2) Generate a full SDR prompt pack
kiessclaw prompt pack outreach_sdr --json-input examples/outreach_sdr.json --out prompts.json --deterministic

# 3) Scaffold a runnable use case folder
kiessclaw scaffold usecase outreach_sdr --json-input examples/outreach_sdr.json --outdir ./generated --force

# 4) Run workflow simulation
kiessclaw workflow run outreach_sdr --json-input examples/outreach_sdr.json --dry-run

# 5) Check providers
kiessclaw providers list
kiessclaw providers health

# 6) Run health diagnostics
kiessclaw doctor

Project Structure

kiessclaw/
├── kiessclaw/
│   ├── core/           # Agent runtime, memory engine, LLM router
│   ├── agents/         # Individual agent implementations
│   ├── skills/         # Modular skill modules
│   ├── models/         # Data models (Contact, Sequence, etc.)
│   ├── usecases/       # Use case schemas and registry
│   ├── prompting/      # Prompt factory + linting
│   ├── codegen/        # Safe scaffold generator + templates
│   ├── seo/            # SEO agents + crawl skills
│   ├── pipeline.py     # Autonomous SEO -> SDR routing
│   ├── interfaces/     # CLI and Slack interfaces
│   └── utils/          # Helpers and utilities
├── config/
│   ├── config.yaml     # Global settings & model assignments
│   └── agents/         # Per-agent directive files (AgentMD)
├── workspace/
│   ├── data/           # JSON data storage
│   ├── memory/         # Persistent agent memory
│   ├── reports/        # Generated reports
│   └── logs/           # Activity logs
├── docs/               # Documentation
├── ROADMAP.md          # Release roadmap milestones
└── scripts/            # Setup and utility scripts

CLI Commands

Command Description
kiessclaw status Show agent status and data summary
kiessclaw usecase list List built-in generator use cases
kiessclaw usecase show <id> Show one use case schema
kiessclaw prompt render <id> --template ... --json-input ... Render one prompt template
kiessclaw prompt pack <id> --json-input ... --out ... Render all prompt templates into one pack
kiessclaw scaffold usecase <id> --json-input ... --outdir ... Generate use case scaffold files
kiessclaw scaffold agent <Name> --outdir ... Generate standalone agent scaffold + test
kiessclaw scaffold workflow <Name> --outdir ... Generate standalone workflow scaffold + test
kiessclaw scaffold skill <Name> --outdir ... Generate standalone skill scaffold + test
kiessclaw scaffold tests <Kind> --outdir ... Generate standalone test stub
kiessclaw workflow run <id> --json-input ... --dry-run Execute a use case workflow end-to-end
kiessclaw providers list Show registered providers and active provider by kind
kiessclaw providers health Run active provider health checks
kiessclaw doctor Check config/workspace/dependencies and show quick fixes
kiessclaw import --file X Import contacts from CSV
kiessclaw contact add ... Add one contact directly from CLI
kiessclaw sequence create --name X Create a new sequence
kiessclaw sequence add-step --id X ... Add sequence step content
kiessclaw enroll <contact> <sequence> Enroll a contact into a sequence
kiessclaw run --dry-run Simulate send + inbox cycle with no state mutation
kiessclaw run Execute one operational cycle
kiessclaw inbox process Process unclassified replies
kiessclaw reply --contact-id X --text "..." Simulate inbound reply
kiessclaw analytics Show pipeline + deliverability summary
kiessclaw pipeline Show prospect pipeline
kiessclaw report Generate weekly report
kiessclaw analyze AI-powered performance analysis
kiessclaw audit --url URL --auto-enroll Run SEO audit and optionally route low-score domains
kiessclaw serve --port 8000 Start FastAPI service layer
kiessclaw chat Interactive agent chat

Configuration

Edit config/config.yaml:

llm:
  provider: openai          # openai | anthropic | ollama
  default_model: gpt-4o
  api_key: ${OPENAI_API_KEY}

agents:
  KPRO:
    model: gpt-4o
    settings:
      min_icp_score: 60
  KSEQ:
    model: gpt-4o-mini
    settings:
      max_daily_sends: 200
      sending_window_start: "08:00"
      sending_window_end: "18:00"

email:
  provider: smtp
  smtp:
    host: ${SMTP_HOST}
    port: 587
    username: ${SMTP_USERNAME}
    password: ${SMTP_PASSWORD}

safety:
  max_sends_per_day: 500
  bounce_rate_threshold: 0.05

Built-in Skills

Skill Description
prospect CSV import, email validation, ICP scoring
sequence Sequence CRUD, enrollment, scheduling
email SMTP sending, tracking, deliverability
personalization Template rendering, first-line generation
inbox Reply classification, DNC management
analytics Funnel metrics, reports, analysis
crawl Technical crawl, metadata, broken links, robots/sitemap checks

Documentation

API Endpoints

  • GET /health
  • GET /version
  • GET /usecases
  • GET /usecases/{id}
  • POST /providers/health
  • POST /prompt/render
  • POST /scaffold/usecase
  • POST /workflow/run
  • POST /audit
  • POST /contacts
  • GET /contacts
  • POST /sequences/{id}/enroll
  • GET /pipeline
  • POST /reply
  • GET /inbox
  • GET /analytics
  • GET /sequences
  • POST /sequences
  • POST /waitlist
  • GET /waitlist/count

Contributing

Contributions are welcome! KIESSCLAW is designed to grow through community extensions.

See CONTRIBUTING.md for guidelines.


License

MIT License — see LICENSE for details.


Author

Erwin Kiess-Alfonso Founder & CEO, KLYTICS · Data & AI Performance Consultancy San Juan, Puerto Rico

Twitter LinkedIn KLYTICS


KIESSCLAW is part of the KLYTICS open-source ecosystem.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

kiessclaw-1.4.0.tar.gz (103.9 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

kiessclaw-1.4.0-py3-none-any.whl (110.6 kB view details)

Uploaded Python 3

File details

Details for the file kiessclaw-1.4.0.tar.gz.

File metadata

  • Download URL: kiessclaw-1.4.0.tar.gz
  • Upload date:
  • Size: 103.9 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.14.2

File hashes

Hashes for kiessclaw-1.4.0.tar.gz
Algorithm Hash digest
SHA256 5154c5d14e362a8b45e2b97f2f690430f5c1333c3c5261e7e39f806401909e72
MD5 4aeecf97dd2a883ff2832bfc84104dd3
BLAKE2b-256 ab2b4d041978aff7fba51c75b6a6f889a1a794cd2e10558c13f4ec17692d0132

See more details on using hashes here.

File details

Details for the file kiessclaw-1.4.0-py3-none-any.whl.

File metadata

  • Download URL: kiessclaw-1.4.0-py3-none-any.whl
  • Upload date:
  • Size: 110.6 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.14.2

File hashes

Hashes for kiessclaw-1.4.0-py3-none-any.whl
Algorithm Hash digest
SHA256 8726a355f56c7bc7f037fccdc783bc8aa41b11c858cac30629dbd85c6564f84f
MD5 1d0ceb04fa303ab6ea7eb56ae3cc9a21
BLAKE2b-256 910e6917be007f7c1311c535164a0e76d221d8b32119bba735f1264ccd22bd83

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page