Is your SaaS product ready for the agent era?
Satya Nadella says the app layer is collapsing into agents. When AI agents become the primary way your product is used, which features survive — and which break completely? Find out before your competitors do.
“The traditional application layer is collapsing into agents.”
— Satya Nadella, Microsoft CEO
The framework
Six dimensions. One clear score.
Developed during the botlington.com experiment (54 commits, 12 board reviews, 6 real product audits). Battle-tested against Carrd, Balsamiq, Kit, Ghost, Plausible, and Linear.
01
API Surface Coverage
Can an agent actually reach your product programmatically? We assess endpoint coverage, documentation quality, OpenAPI specs, and rate limit design.
02
Programmatic Authentication
Can software authenticate without a human click-through? API keys, OAuth client credentials, scoped PATs, service accounts — what works and what's missing.
03
Structured Data Output
Does your API return structured data, or are agents left scraping HTML? We review response formats, webhook support, and machine-readable error handling.
04
MCP & Tool Interface
Is there a clean integration layer for agents? MCP server, function-calling SDK, or dedicated agent tooling — we assess discoverability and usability.
05
Permissions & Safety
Can you scope what an agent is allowed to do? Least-privilege tokens, read/write/delete granularity, expiry policies, and audit trails.
06
Agent Observability
Can you see agent traffic and debug it? We review logging, agent identification, usage analytics, and anomaly detection capabilities.
What you get
€299
One-time. No subscription. No upsell.
- Full 6-dimension assessment with /10 scoring per dimension
- Written Agent Readiness Report (PDF, shareable with your team)
- Prioritised remediation roadmap — what to fix first for maximum impact
- 60-minute strategy session to walk through findings and answer questions
- 30-day email follow-up for implementation questions
Not useful? Full refund. No questions asked.
Delivered by
Phil Bennett
The AI built the framework. Phil delivers the expertise. Here's why that matters: the botlington.com experiment proved that AI can build rigorous assessment tools — and that humans are still required for trust, nuance, and the judgment calls that turn a score into a strategy.
- 20 years engineering leadership — CTO, VP Engineering, Head of Platform
- Ex-Klarna (payments infrastructure at scale)
- Ex-Corvus Insurance (data platform, API architecture)
- Fractional CTO for multiple SaaS startups
- Author of Punk Leadership (Substack) on engineering culture
- Builder of OpenClaw — open-source AI agent framework
Proof of rigour
6 products already scored.
Read the full sample audits. This is the same framework and methodology applied to your product, with Phil's 20 years of architecture experience on top.
0/10
Carrd
1/10
Balsamiq
6/10
Kit
7/10
Ghost
7/10
Plausible
7/10
Linear
Want to start smaller?
Get the Agent Readiness Audit first — pay what you want, from €1. Same 6-dimension framework, delivered as a PDF + Loom walkthrough within 48 hours. Upgrade to full consulting any time.