Sun, Nov 8, 2026
3 notes this week
The way agents ‘read’ code is different - good agents are all about gathering data about how the code works. They’ll prod around with it, execute parts of it.
So ‘readability’ to agents really means ‘verifiability’. How easy it is to mess about with that code to see it working.
Let’s take this scenario:
You have a huge module in your codebase that uses all sorts of complex performance tricks
It has 100% test coverage
A talented human would need an hour or two to get comfortable working in that module.
An agent needs zero time. They verify it with the tests, poke about a bit, and they’re good to go.
— Matt Pocock, on code readability in the age of agents Experts go with their AI agent's answer
Anthropic gave 50 data analysts a Claude Code agent and a realistic analysis with four open decisions. Each decision had two reasonable answers. In secret, a coin flip told the agent which answer to argue for. It had to argue honestly.
The analysts picked the agent’s answer 86% of the time. Afterwards, 77% explained their answers with the agent’s arguments, sometimes word for word. Only 15% gave the agent credit. Analysts who said “it was my decision” picked the agent’s answer just as often as the rest.
- The coin flips alone predicted 82% of the final recommendations. The agent didn’t need to argue for the recommendation. It had already picked the inputs.
- With no instructions to argue for anything, Opus 4.8 gave the same five answers in 112 of 123 runs. Ten teams on one model agreeing is one model agreeing with itself.
This is the limit of my meat proxy note. Rewriting AI output in your own words proves you read it. It doesn’t prove the decision was yours. For the two or three decisions that matter, write down your answer before you ask the agent.
pi-schema
pi-schema is a JSON Schema validator from the Pi monorepo that never calls
evalornew Function. Mario Zechner says he built it with a single prompt. It compiles each schema into plain closures, so it stays fast in places that block code generation.- The need is real. Cloudflare Workers block
eval()andnew Function. Chrome MV3 extensions cannot allow'unsafe-eval'. Ajv, the default JSON Schema validator in JS, compiles schemas withnew Function. It broke MCP elicitation on Workers (typescript-sdk#689). - How people cope today. Ajv can generate validators at build time, but that only works for schemas you know in advance. The MCP SDK lists @cfworker/json-schema, an interpreter, as a peer dependency. Zod and ArkType have
jitlessflags, and TypeBox falls back to its interpreter. Zod’s flag has to be set before any import that uses Zod (zod#5789). - The numbers. Without eval, the README puts a nested schema at 407 ns for pi-schema, 42 µs for TypeBox and 43 µs for cfworker. pi-schema passes 2,222 of the 2,223 core tests in the official suite.
Agents make this matter. An agent learns its tools from MCP servers while it runs:
tools/listreturns each tool’s JSON Schema. You cannot compile those ahead of time, and the harnesses are moving to the edge. Pi’s Chord design notes plan a Cloudflare Durable Object driver, and Pi Durable already stores agent work as checkpointed tasks. The catches: these are the author’s own benchmarks, Ajv is not in the comparison, the package is not on npm yet, and TypeBox’s JIT still wins on small flat schemas (9 ns vs 31 ns).- The need is real. Cloudflare Workers block