2026-02-20 10:32:07 -06:00
|
|
|
import { beforeEach, describe, expect, it, vi } from "vitest";
|
2026-03-03 08:45:26 -06:00
|
|
|
import { models as codexFallbackModels } from "@paperclipai/adapter-codex-local";
|
2026-03-05 07:52:23 -06:00
|
|
|
import { models as cursorFallbackModels } from "@paperclipai/adapter-cursor-local";
|
2026-03-05 18:59:42 -06:00
|
|
|
import { models as opencodeFallbackModels } from "@paperclipai/adapter-opencode-local";
|
2026-03-05 15:24:20 +01:00
|
|
|
import { resetOpenCodeModelsCacheForTests } from "@paperclipai/adapter-opencode-local/server";
|
[codex] Improve transient recovery and Codex model refresh (#4383)
## Thinking Path
> - Paperclip orchestrates AI agents for zero-human companies
> - Adapter execution and retry classification decide whether agent work
pauses, retries, or recovers automatically
> - Transient provider failures need to be classified precisely so
Paperclip does not convert retryable upstream conditions into false hard
failures
> - At the same time, operators need an up-to-date model list for
Codex-backed agents and prompts should nudge agents toward targeted
verification instead of repo-wide sweeps
> - This pull request tightens transient recovery classification for
Claude and Codex, updates the agent prompt guidance, and adds Codex
model refresh support end-to-end
> - The benefit is better automatic retry behavior plus fresher
operator-facing model configuration
## What Changed
- added Codex usage-limit retry-window parsing and Claude extra-usage
transient classification
- normalized the heartbeat transient-recovery contract across adapter
executions and heartbeat scheduling
- documented that deferred comment wakes only reopen completed issues
for human/comment-reopen interactions, while system follow-ups leave
closed work closed
- updated adapter-utils prompt guidance to prefer targeted verification
- added Codex model refresh support in the server route, registry,
shared types, and agent config form
- added adapter/server tests covering the new parsing, retry scheduling,
and model-refresh behavior
## Verification
- `pnpm exec vitest run --project @paperclipai/adapter-utils
packages/adapter-utils/src/server-utils.test.ts`
- `pnpm exec vitest run --project @paperclipai/adapter-claude-local
packages/adapters/claude-local/src/server/parse.test.ts`
- `pnpm exec vitest run --project @paperclipai/adapter-codex-local
packages/adapters/codex-local/src/server/parse.test.ts`
- `pnpm exec vitest run --project @paperclipai/server
server/src/__tests__/adapter-model-refresh-routes.test.ts
server/src/__tests__/adapter-models.test.ts
server/src/__tests__/claude-local-execute.test.ts
server/src/__tests__/codex-local-execute.test.ts
server/src/__tests__/heartbeat-process-recovery.test.ts
server/src/__tests__/heartbeat-retry-scheduling.test.ts`
## Risks
- Moderate behavior risk: retry classification affects whether runs
auto-recover or block, so mistakes here could either suppress needed
retries or over-retry real failures
- Low workflow risk: deferred comment wake reopening is intentionally
scoped to human/comment-reopen interactions so system follow-ups do not
revive completed issues unexpectedly
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex GPT-5-based coding agent with tool use and code execution
in the Codex CLI environment
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [ ] If this change affects the UI, I have included before/after
screenshots
- [ ] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-04-24 09:40:40 -05:00
|
|
|
import { listAdapterModels, refreshAdapterModels } from "../adapters/index.js";
|
2026-02-20 10:32:07 -06:00
|
|
|
import { resetCodexModelsCacheForTests } from "../adapters/codex-models.js";
|
2026-03-05 07:52:23 -06:00
|
|
|
import { resetCursorModelsCacheForTests, setCursorModelsRunnerForTests } from "../adapters/cursor-models.js";
|
2026-02-20 10:32:07 -06:00
|
|
|
|
|
|
|
|
describe("adapter model listing", () => {
|
|
|
|
|
beforeEach(() => {
|
|
|
|
|
delete process.env.OPENAI_API_KEY;
|
2026-03-05 15:24:20 +01:00
|
|
|
delete process.env.PAPERCLIP_OPENCODE_COMMAND;
|
2026-02-20 10:32:07 -06:00
|
|
|
resetCodexModelsCacheForTests();
|
2026-03-05 07:52:23 -06:00
|
|
|
resetCursorModelsCacheForTests();
|
|
|
|
|
setCursorModelsRunnerForTests(null);
|
2026-03-05 15:24:20 +01:00
|
|
|
resetOpenCodeModelsCacheForTests();
|
2026-02-20 10:32:07 -06:00
|
|
|
vi.restoreAllMocks();
|
|
|
|
|
});
|
|
|
|
|
|
|
|
|
|
it("returns an empty list for unknown adapters", async () => {
|
|
|
|
|
const models = await listAdapterModels("unknown_adapter");
|
|
|
|
|
expect(models).toEqual([]);
|
|
|
|
|
});
|
|
|
|
|
|
|
|
|
|
it("returns codex fallback models when no OpenAI key is available", async () => {
|
|
|
|
|
const fetchSpy = vi.spyOn(globalThis, "fetch");
|
|
|
|
|
const models = await listAdapterModels("codex_local");
|
|
|
|
|
|
|
|
|
|
expect(models).toEqual(codexFallbackModels);
|
|
|
|
|
expect(fetchSpy).not.toHaveBeenCalled();
|
|
|
|
|
});
|
|
|
|
|
|
|
|
|
|
it("loads codex models dynamically and merges fallback options", async () => {
|
|
|
|
|
process.env.OPENAI_API_KEY = "sk-test";
|
|
|
|
|
const fetchSpy = vi.spyOn(globalThis, "fetch").mockResolvedValue({
|
|
|
|
|
ok: true,
|
|
|
|
|
json: async () => ({
|
|
|
|
|
data: [
|
|
|
|
|
{ id: "gpt-5-pro" },
|
|
|
|
|
{ id: "gpt-5" },
|
|
|
|
|
],
|
|
|
|
|
}),
|
|
|
|
|
} as Response);
|
|
|
|
|
|
|
|
|
|
const first = await listAdapterModels("codex_local");
|
|
|
|
|
const second = await listAdapterModels("codex_local");
|
|
|
|
|
|
|
|
|
|
expect(fetchSpy).toHaveBeenCalledTimes(1);
|
|
|
|
|
expect(first).toEqual(second);
|
|
|
|
|
expect(first.some((model) => model.id === "gpt-5-pro")).toBe(true);
|
|
|
|
|
expect(first.some((model) => model.id === "codex-mini-latest")).toBe(true);
|
|
|
|
|
});
|
|
|
|
|
|
[codex] Improve transient recovery and Codex model refresh (#4383)
## Thinking Path
> - Paperclip orchestrates AI agents for zero-human companies
> - Adapter execution and retry classification decide whether agent work
pauses, retries, or recovers automatically
> - Transient provider failures need to be classified precisely so
Paperclip does not convert retryable upstream conditions into false hard
failures
> - At the same time, operators need an up-to-date model list for
Codex-backed agents and prompts should nudge agents toward targeted
verification instead of repo-wide sweeps
> - This pull request tightens transient recovery classification for
Claude and Codex, updates the agent prompt guidance, and adds Codex
model refresh support end-to-end
> - The benefit is better automatic retry behavior plus fresher
operator-facing model configuration
## What Changed
- added Codex usage-limit retry-window parsing and Claude extra-usage
transient classification
- normalized the heartbeat transient-recovery contract across adapter
executions and heartbeat scheduling
- documented that deferred comment wakes only reopen completed issues
for human/comment-reopen interactions, while system follow-ups leave
closed work closed
- updated adapter-utils prompt guidance to prefer targeted verification
- added Codex model refresh support in the server route, registry,
shared types, and agent config form
- added adapter/server tests covering the new parsing, retry scheduling,
and model-refresh behavior
## Verification
- `pnpm exec vitest run --project @paperclipai/adapter-utils
packages/adapter-utils/src/server-utils.test.ts`
- `pnpm exec vitest run --project @paperclipai/adapter-claude-local
packages/adapters/claude-local/src/server/parse.test.ts`
- `pnpm exec vitest run --project @paperclipai/adapter-codex-local
packages/adapters/codex-local/src/server/parse.test.ts`
- `pnpm exec vitest run --project @paperclipai/server
server/src/__tests__/adapter-model-refresh-routes.test.ts
server/src/__tests__/adapter-models.test.ts
server/src/__tests__/claude-local-execute.test.ts
server/src/__tests__/codex-local-execute.test.ts
server/src/__tests__/heartbeat-process-recovery.test.ts
server/src/__tests__/heartbeat-retry-scheduling.test.ts`
## Risks
- Moderate behavior risk: retry classification affects whether runs
auto-recover or block, so mistakes here could either suppress needed
retries or over-retry real failures
- Low workflow risk: deferred comment wake reopening is intentionally
scoped to human/comment-reopen interactions so system follow-ups do not
revive completed issues unexpectedly
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex GPT-5-based coding agent with tool use and code execution
in the Codex CLI environment
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [ ] If this change affects the UI, I have included before/after
screenshots
- [ ] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-04-24 09:40:40 -05:00
|
|
|
it("refreshes cached codex models on demand", async () => {
|
|
|
|
|
process.env.OPENAI_API_KEY = "sk-test";
|
|
|
|
|
const fetchSpy = vi.spyOn(globalThis, "fetch")
|
|
|
|
|
.mockResolvedValueOnce({
|
|
|
|
|
ok: true,
|
|
|
|
|
json: async () => ({
|
|
|
|
|
data: [{ id: "gpt-5" }],
|
|
|
|
|
}),
|
|
|
|
|
} as Response)
|
|
|
|
|
.mockResolvedValueOnce({
|
|
|
|
|
ok: true,
|
|
|
|
|
json: async () => ({
|
|
|
|
|
data: [{ id: "gpt-5.5" }],
|
|
|
|
|
}),
|
|
|
|
|
} as Response);
|
|
|
|
|
|
|
|
|
|
const initial = await listAdapterModels("codex_local");
|
|
|
|
|
const refreshed = await refreshAdapterModels("codex_local");
|
|
|
|
|
|
|
|
|
|
expect(fetchSpy).toHaveBeenCalledTimes(2);
|
|
|
|
|
expect(initial.some((model) => model.id === "gpt-5")).toBe(true);
|
|
|
|
|
expect(refreshed.some((model) => model.id === "gpt-5.5")).toBe(true);
|
|
|
|
|
});
|
|
|
|
|
|
2026-02-20 10:32:07 -06:00
|
|
|
it("falls back to static codex models when OpenAI model discovery fails", async () => {
|
|
|
|
|
process.env.OPENAI_API_KEY = "sk-test";
|
|
|
|
|
vi.spyOn(globalThis, "fetch").mockResolvedValue({
|
|
|
|
|
ok: false,
|
|
|
|
|
status: 401,
|
|
|
|
|
json: async () => ({}),
|
|
|
|
|
} as Response);
|
|
|
|
|
|
|
|
|
|
const models = await listAdapterModels("codex_local");
|
|
|
|
|
expect(models).toEqual(codexFallbackModels);
|
|
|
|
|
});
|
2026-03-05 07:52:23 -06:00
|
|
|
|
2026-03-06 15:23:55 +00:00
|
|
|
|
2026-03-05 07:52:23 -06:00
|
|
|
it("returns cursor fallback models when CLI discovery is unavailable", async () => {
|
|
|
|
|
setCursorModelsRunnerForTests(() => ({
|
|
|
|
|
status: null,
|
|
|
|
|
stdout: "",
|
|
|
|
|
stderr: "",
|
|
|
|
|
hasError: true,
|
|
|
|
|
}));
|
|
|
|
|
|
|
|
|
|
const models = await listAdapterModels("cursor");
|
|
|
|
|
expect(models).toEqual(cursorFallbackModels);
|
|
|
|
|
});
|
|
|
|
|
|
2026-03-05 18:59:42 -06:00
|
|
|
it("returns opencode fallback models including gpt-5.4", async () => {
|
2026-03-08 12:10:59 -05:00
|
|
|
process.env.PAPERCLIP_OPENCODE_COMMAND = "__paperclip_missing_opencode_command__";
|
|
|
|
|
|
2026-03-05 18:59:42 -06:00
|
|
|
const models = await listAdapterModels("opencode_local");
|
|
|
|
|
|
|
|
|
|
expect(models).toEqual(opencodeFallbackModels);
|
|
|
|
|
});
|
|
|
|
|
|
2026-03-05 07:52:23 -06:00
|
|
|
it("loads cursor models dynamically and caches them", async () => {
|
|
|
|
|
const runner = vi.fn(() => ({
|
|
|
|
|
status: 0,
|
|
|
|
|
stdout: "Available models: auto, composer-1.5, gpt-5.3-codex-high, sonnet-4.6",
|
|
|
|
|
stderr: "",
|
|
|
|
|
hasError: false,
|
|
|
|
|
}));
|
|
|
|
|
setCursorModelsRunnerForTests(runner);
|
|
|
|
|
|
|
|
|
|
const first = await listAdapterModels("cursor");
|
|
|
|
|
const second = await listAdapterModels("cursor");
|
|
|
|
|
|
|
|
|
|
expect(runner).toHaveBeenCalledTimes(1);
|
|
|
|
|
expect(first).toEqual(second);
|
|
|
|
|
expect(first.some((model) => model.id === "auto")).toBe(true);
|
|
|
|
|
expect(first.some((model) => model.id === "gpt-5.3-codex-high")).toBe(true);
|
|
|
|
|
expect(first.some((model) => model.id === "composer-1")).toBe(true);
|
|
|
|
|
});
|
2026-03-06 15:23:55 +00:00
|
|
|
|
2026-02-20 10:32:07 -06:00
|
|
|
});
|