Put Piggy on Prime Agent, and let it write to the book

Piggy was a hand-rolled OpenAI tool loop. It is now a Prime Agent session —
Prime Intellect's own harness, embedded as a Node library — answering from
PIG's tools and, for the first time, able to put information into the CRM
rather than only read it out.

The harness is a coding agent, so the first job was taking the coding agent
away from it. `noTools: 'all'` plus an explicit allowlist leaves the model
with PIG's ten `pig_*` tools and no bash, no filesystem, no IPython. That
holds under attack: a hostile extension, a skill and a settings file planted
in the agent's own directory, then `setActiveToolsByName` called with every
built-in, still leaves ten tools, all ours. Both lines are load-bearing —
`noTools` alone registers nothing, and the allowlist is what admits our own.

Writing is gated rather than assumed. A change is proposed, not made: the
tool returns a description, the transcript renders a diff card, and nothing
reaches the database until someone presses Apply. Contracts, commitments,
allocations and compliance always stop for a human whatever the mode. Every
write runs through `executeMutation` as the calling user, so their
capabilities and the audit trail apply exactly as they would to a human's.

Four things about the SDK are wrong in its own documentation and cost a
debugging cycle each: models.json does not resolve an env var name for
`apiKey`, it sends the literal string; there is no built-in prime-inference
provider in 0.84.1; a ResourceLoader you pass in is never reloaded for you;
and the stock system prompt is a coding-assistant prompt that must be
replaced — but replacing it also silently removes the tool list, because the
harness only renders that section when it owns the prompt. AGENTS.md records
all four.

The expensive one was thinking level. The harness defaults to `medium`, and
nemotron spent an entire 4,096-token budget reasoning and returned an empty
answer. `low` was worse; `off` omits the parameter so the endpoint's default
wins. An explicit `reasoning_effort: none` via `thinkingLevelMap` took a turn
from 6,195 output tokens to 149.

And a turn is now bounded. The harness loop is `while (true)` with no
iteration cap; a runaway on a frontier model would have eaten the credit it
is supposed to report on. Ceilings on model calls and tokens, enforced both
through the harness hook and independently from the event stream, plus a
per-user daily spend limit — and the ledger now records spend on turns that
fail, which it previously discarded.

Signing in lands on /piggy, which is a workspace: conversations down one
side, the agent in the middle, what it did and what it cost beside it.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

apps/piggy/src/dev/verify-prime-agent.ts

@@ -0,0 +1,86 @@
+/**
+ * Proves the Prime Agent runtime against the real endpoint.
+ *
+ * A typecheck cannot tell you that the credential resolved, that the loader was
+ * reloaded, or that no built-in tool survived `noTools: 'all'` — every one of
+ * those failures compiles perfectly and shows up as a 401, a coding-assistant
+ * answer, or a shell in a CRM. So this asks the live model a question with a
+ * seeded tool behind it and prints what actually happened.
+ *
+ *   corepack pnpm -F @pig/piggy exec tsx src/dev/verify-prime-agent.ts [modelId]
+ *
+ * Requires PRIME_API_KEY. It spends a few hundred tokens; it is a dev tool, not
+ * a test, and nothing in CI runs it.
+ */
+import { defineTool } from '@earendil-works/pi-coding-agent';
+import { Type } from 'typebox';
+import { createPiggySession } from '../agent/session';
+
+const tool = defineTool({
+  name: 'pig_get_workspace_summary',
+  label: 'Workspace summary',
+  description: 'Returns the workspace-wide capacity aggregates, already computed.',
+  promptSnippet: 'pig_get_workspace_summary: workspace-wide capacity aggregates, already computed.',
+  parameters: Type.Object({}),
+  async execute() {
+    console.log('    [tool] pig_get_workspace_summary called');
+    return {
+      content: [
+        {
+          type: 'text' as const,
+          // The figures are chosen to catch the two failures that matter: 189
+          // must be read as $1.89 and 112 as $1.12, not as "189" and "112
+          // cents".
+          text: JSON.stringify({
+            headline: 'Northwind Robotics H100 block, 38% sold',
+            committedGpuHours: 52_000,
+            allocatedGpuHours: 19_760,
+            utilisation: 0.38,
+            costPerGpuHourCents: 189,
+            breakEvenPriceCents: 112,
+            idleCostCents: 1_200_000,
+          }),
+        },
+      ],
+      details: {},
+    };
+  },
+});
+
+const modelId = process.argv[2];
+const piggy = await createPiggySession({
+  mode: 'confirm',
+  ...(modelId ? { modelId } : {}),
+  tools: [tool],
+});
+
+const live = piggy.session.agent.state.tools.map((entry) => entry.name);
+const shellish = live.filter((name) =>
+  /^(bash|shell|ipython|python|read|write|edit|ls|grep|find)$/i.test(name),
+);
+
+console.log('MODEL:', piggy.modelId);
+console.log('TOOLS:', live);
+console.log('SHELL/PYTHON PRESENT:', shellish.length > 0);
+console.log('SYSTEM PROMPT (first 200):', piggy.session.systemPrompt.slice(0, 200));
+console.log('PROMPT LISTS THE TOOL:', piggy.session.systemPrompt.includes('pig_get_workspace_summary'));
+console.log('PROMPT IS THE CODING PREAMBLE:', /coding assistant/i.test(piggy.session.systemPrompt));
+console.log('---');
+
+let answer = '';
+const unsubscribe = piggy.session.subscribe((event) => {
+  if (event.type === 'message_update' && event.assistantMessageEvent.type === 'text_delta') {
+    answer += event.assistantMessageEvent.delta;
+  }
+  if (event.type === 'tool_execution_start') console.log('    [event] tool_execution_start');
+});
+
+await piggy.session.prompt(
+  'What is the break-even price per GPU-hour on this block, and how much has the idle capacity already cost? Use the tool.',
+);
+await piggy.session.waitForIdle();
+unsubscribe();
+
+console.log('ANSWER:', answer.trim());
+piggy.dispose();
+process.exit(0);