Commit Graph
6 Commits
Author SHA1 Message Date
dckcandClaude Opus 5.5 4a85143a66 fix(good-fences): give the model room to answer
gpt-oss-120b reasons before it answers, and Workers AI's default
max_tokens (256) ran out mid-thought: finish_reason "length", content
null. Ask for 4096, and when there's no reply text, say which choice
and message keys came back instead.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-10-09 00:21:44 -05:00
dckcandClaude Opus 5.5 37faa92649 test(good-fences): lent a mower, the model writes code to mow (red)
Code mode (2c): the harness turns what I say plus `about` into a prompt;
the model replies with `(env) => ...` code. Checked textually for now,
since workerd won't eval it: the code pushes `env.mower` and calls only
methods in `about`. It also prints the code, to see what the model writes.

Red: src/code-mode.ts doesn't exist yet.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-10-09 00:01:39 -05:00
dckcandClaude Opus 5.5 ec24f3e8e4 feat(good-fences): complete(messages) on a Workers AI model
makeWorkersAiModel narrows the AI binding to one model's chat
completion: the harness gets `complete`, not `run` on any model at
any cost. gpt-oss-120b replies in the OpenAI shape (checked with a
one-off probe), so the text is choices[0].message.content; anything
else throws, naming the keys it got.

Green: the model counts 3 rs in strawberry.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-10-08 23:41:18 -05:00
dckcandClaude Opus 5.5 47a2337afe test(good-fences): the model counts the rs in strawberry (red)
A first integration test of the model call: Workers AI through the AI
binding, behind the same `complete(messages)` interface as the sim's
fake model. Opt-in (`pnpm test:model`) since it needs `wrangler login`
and costs money.

Red: src/model.ts doesn't exist yet.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-10-08 23:36:46 -05:00
dckcandClaude Opus 5.5 9b12943be4 test(good-fences): show which mower the neighbor gets
The chat and load arrows drew `things: { mower }` whether I lent the
raw mower or the governed one. Name objects by the reply arrow that
introduced them, so the diagram reads `things: { mower: governed }`
and lending the raw mower would fail the trace comparison.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-10-08 23:27:13 -05:00
dckcandClaude Opus 5.5 9a3e7348c6 test(good-fences): lend-the-mower journey (red)
The spec is a sequence diagram of the a2030.md story: I lend a governed
mower; the neighbor's harness shows the model only `about`, runs its
code as a tenant, and the governor stops the tipsy tenant at a 2-cell
fence. The test runs the actors through a causal tracer and compares
the trace with the diagram, then checks the yard and what the model saw.

The tracer comes from agoric-sdk PR #12895, extended to name
participants made mid-sequence (governor, tenant) and to draw thrown
errors as replies.

Red: src/ doesn't exist yet (ERR_MODULE_NOT_FOUND for src/fence.ts).

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-10-08 22:17:48 -05:00