Skip to content

ci(pullfrog): run gpt-6.1-sol through CLIProxy again - #34

Merged
onmax merged 5 commits into
mainfrom
ci/pullfrog-cliproxy
Nov 10, 2026
Merged

onmax merged 5 commits into
mainfrom
ci/pullfrog-cliproxy

Conversation

@onmax

@onmax onmax commented Nov 10, 2026 •

Copy link
Copy Markdown
Owner

Every Pullfrog run here fails before the agent starts: the Codex subscription stored in Pullfrog hit its usage limit and resets on 13 November. It's the only credential connected, so there's nothing to fall back to.

This goes back to CLIProxy, as before #33: openai-compatible/byok at https://ai.onmax.me/v1 with gpt-6.1-sol and the existing CLIPROXY_* secrets. The Cloudflare Access headers now go through OPENAI_COMPATIBLE_HEADERS, which Pullfrog v0.1.123 reads itself, so the OpenCode config step is gone. A completion check runs before the agent, so a proxy failure shows up as its own step. Same change as onmax/nuxt-skill-hub#54.

I dispatched the workflow on this branch with a one-word prompt: run 42785358087. OpenCode declares only low to medium effort for the proxy model, so effort: xhigh alone lands on medium. The model is gpt-6.1-sol(xhigh) instead: CLIProxy applies that suffix over the request's reasoning_effort. A test prompt sent with reasoning_effort: low used 3,366 reasoning tokens with the suffix and 1,016 without. This run on onmax/tiny shows openai-compatible/gpt-6.1-sol(xhigh).

The Codex subscription stored in Pullfrog hit its usage limit, so every
run fails before the agent starts. CLIProxy answers again, so runs go
back to it. The Access headers now go through OPENAI_COMPATIBLE_HEADERS,
which Pullfrog v0.1.123 reads, instead of an OpenCode config file.
@pullfrog

pullfrog Bot commented Nov 10, 2026 •

Copy link
Copy Markdown

No configured credential can serve this run. Your ChatGPT subscription has hit its usage limit, which resets at 2026-11-13 17:11 UTC. Re-trigger after it resets, or add an API key for this provider, which Pullfrog falls back to automatically.

The providers' responses
CODEX_AUTH_JSON: the ChatGPT plan's usage limit is reached

Pick another model →

Pullfrog  | Rerun failed job ➔ | View workflow run | via Pullfrog | Using gpt-6.1-sol

@chatgpt-codex-connector

Copy link
Copy Markdown
Codex Review Summary: ❌ Failed

This comment shows the latest Codex review activity on this pull request.

Review Status Commit Review trigger
📝 Code Review ❌ Failed 2026-11-10T08:49:20.852575Z
We couldn’t complete this review. Comment @codex review to try again.
338bde1 PR opened
ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

@chatgpt-codex-connector

Copy link
Copy Markdown

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.

@pkg-pr-new

pkg-pr-new Bot commented Nov 10, 2026 •

Copy link
Copy Markdown

Open in StackBlitz

pnpm add https://pkg.pr.new/tinyaml@34
pnpm add https://pkg.pr.new/tinyini@34
pnpm add https://pkg.pr.new/tinyjson5@34
pnpm add https://pkg.pr.new/tinyswap@34

commit: 1f879fa

Pullfrog lands the effort input on the variants OpenCode declares for the
model, so without it runs fall back to medium.
@pullfrog

pullfrog Bot commented Nov 10, 2026

Copy link
Copy Markdown

No configured credential can serve this run. Your ChatGPT subscription has hit its usage limit, which resets at 2026-11-13 17:11 UTC. Re-trigger after it resets, or add an API key for this provider, which Pullfrog falls back to automatically.

The providers' responses
CODEX_AUTH_JSON: the ChatGPT plan's usage limit is reached

Pick another model →

Pullfrog  | Rerun failed job ➔ | View workflow run | via Pullfrog | Using gpt-6.1-sol

OpenCode declares only low to medium variants for the proxy model, so
the effort input lands on medium. CLIProxy applies a (xhigh) suffix over
the request's reasoning_effort.
@pullfrog

pullfrog Bot commented Nov 10, 2026

Copy link
Copy Markdown

No configured credential can serve this run. Your ChatGPT subscription has hit its usage limit, which resets at 2026-11-13 17:11 UTC. Re-trigger after it resets, or add an API key for this provider, which Pullfrog falls back to automatically.

The providers' responses
CODEX_AUTH_JSON: the ChatGPT plan's usage limit is reached

Pick another model →

Pullfrog  | Rerun failed job ➔ | View workflow run | via Pullfrog | Using gpt-6.1-sol

@pullfrog

pullfrog Bot commented Nov 10, 2026

Copy link
Copy Markdown

No configured credential can serve this run. Your ChatGPT subscription has hit its usage limit, which resets at 2026-11-13 17:11 UTC. Re-trigger after it resets, or add an API key for this provider, which Pullfrog falls back to automatically.

The providers' responses
CODEX_AUTH_JSON: the ChatGPT plan's usage limit is reached

Pick another model →

Pullfrog  | Rerun failed job ➔ | View workflow run | via Pullfrog | Using gpt-6.1-sol

@pullfrog

pullfrog Bot commented Nov 10, 2026

Copy link
Copy Markdown

No configured credential can serve this run. Your ChatGPT subscription has hit its usage limit, which resets at 2026-11-13 17:11 UTC. Re-trigger after it resets, or add an API key for this provider, which Pullfrog falls back to automatically.

The providers' responses
CODEX_AUTH_JSON: the ChatGPT plan's usage limit is reached

Pick another model →

Pullfrog  | Rerun failed job ➔ | View workflow run | via Pullfrog | Using gpt-6.1-sol

@onmax
onmax merged commit e286229 into main Nov 10, 2026
9 of 10 checks passed
@onmax
onmax deleted the ci/pullfrog-cliproxy branch November 10, 2026 09:42
@github-actions github-actions Bot mentioned this pull request Nov 9, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant