mirror of
https://github.com/nexu-io/open-design.git
synced 2026-06-01 03:14:35 +07:00
c5e38bbe58
* fix(daemon): remove 10-item cap from discovery TodoWrite plan prompt The RULE 3 sentence in DISCOVERY_AND_PHILOSOPHY told the model to write 'a plan of 5–10 short imperative items'. That upper bound caused the agent to cap every plan at exactly ten steps even when the task genuinely needed more. The TodoWrite JSON schema imposes no maxItems constraint, so the cap was entirely prompt-driven. Replace '5–10 short imperative items' with 'short imperative items covering the work'. TodoWrite intent, RULE 3 label, and planning-before-building requirement all survive unchanged. Red spec: apps/daemon/tests/prompts/discovery-todo-cap.test.ts * fix(prompts): remove 10-item cap from contracts discovery copy and harden tests [pass-6,7 BLOCKER] packages/contracts/src/prompts/discovery.ts still had the old '5-10 short imperative items' wording. apps/web imports composeSystemPrompt from @open-design/contracts (ProjectView.tsx:43), so web-originated chat runs were still subject to the cap. [pass-8 WARNING] discovery-todo-cap.test.ts did not cover the contracts copy, leaving that path unguarded. Also no guard against semantically equivalent re-introduction via 'at most / maximum / no more than'. Changes: - packages/contracts/src/prompts/discovery.ts: apply same wording fix as apps/daemon; add inline rationale comment - apps/daemon/src/prompts/discovery.ts: add inline rationale comment - apps/daemon/tests/prompts/discovery-todo-cap.test.ts: add 4th assertion blocking 'at most|maximum|no more than N item' re-introduction - packages/contracts/tests/system-prompt.test.ts: add 5-assertion suite guarding the contracts copy and composed prompt output
38 lines
1.8 KiB
TypeScript
38 lines
1.8 KiB
TypeScript
import { describe, expect, it } from 'vitest';
|
||
|
||
import { DISCOVERY_AND_PHILOSOPHY } from '../../src/prompts/discovery.js';
|
||
|
||
// The system prompt historically told the model to write "a plan of 5–10 short
|
||
// imperative items". That upper bound caused the agent to cap every plan at
|
||
// exactly ten steps and then stop or skip additional items — even when the task
|
||
// genuinely needed more. There is no maxItems constraint in the upstream
|
||
// TodoWrite JSON schema (the array is unbounded), so the cap is entirely
|
||
// prompt-driven and can be removed here.
|
||
//
|
||
// This test locks the absence of the cap so a future prompt edit cannot
|
||
// accidentally re-introduce the "5–10" or "5 to 10" wording.
|
||
|
||
describe('discovery.ts RULE 3 — TodoWrite plan item count', () => {
|
||
it('does not cap the plan at 10 items via "5–10" wording', () => {
|
||
// The old wording was "a plan of 5–10 short imperative items".
|
||
// After the fix the sentence must not mention an upper bound of 10.
|
||
expect(DISCOVERY_AND_PHILOSOPHY).not.toMatch(/5[–\-]10\s+short\s+imperative/);
|
||
});
|
||
|
||
it('does not cap the plan at 10 items via "5 to 10" wording', () => {
|
||
expect(DISCOVERY_AND_PHILOSOPHY).not.toMatch(/5 to 10\s+(?:short\s+)?items/i);
|
||
});
|
||
|
||
it('does not re-introduce a numeric cap via "at most / maximum / no more than" phrasing', () => {
|
||
// Guard against semantically equivalent upper-bound re-introduction.
|
||
expect(DISCOVERY_AND_PHILOSOPHY).not.toMatch(
|
||
/(?:at most|maximum|no more than)\s+1[0-9]\s+(?:todo|plan|step|item)/i,
|
||
);
|
||
});
|
||
|
||
it('still instructs the agent to write at least a few items', () => {
|
||
// The intent — plan with TodoWrite before building — must survive the fix.
|
||
expect(DISCOVERY_AND_PHILOSOPHY).toContain('TodoWrite');
|
||
expect(DISCOVERY_AND_PHILOSOPHY).toContain('RULE 3');
|
||
});
|
||
});
|