examples-auto-run

Runs repository Python examples in auto mode with logging, rerun helpers, and background control.

Updated May 23, 2026
One-click install
npx skills add https://github.com/kiranimmadi2/promptforge-ai --skill examples-auto-run-kiranimmadi2
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: examples-auto-run
Source: https://github.com/kiranimmadi2/promptforge-ai/tree/main/openai-agents-python/.agents/skills/examples-auto-run
Command: npx skills add https://github.com/kiranimmadi2/promptforge-ai --skill examples-auto-run-kiranimmadi2

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve? Running a large suite of Python examples manually is slow and error-prone, especially when examples require interactive input, optional dependency extras, or background services. This Skill automates the full run cycle with consistent logging and failure tracking. ## Core Features & Use Cases - Automated Example Execution: Runs uv run examples/run_examples.py in auto mode with auto-approvals and optional dependency extras (litellm, sqlalchemy, redis, modal, temporal, and more) enabled by default. - Lifecycle Helpers: Provides start, stop, status, logs, tail, collect, and rerun commands via run.sh, including background execution with a pidfile. - Failure Rerun Workflow: Writes failed examples to .tmp/examples-rerun.txt so only failures can be rerun without repeating the full suite. - Use Case: After modifying the SDK, start a full auto-mode run in the background, tail the main log, then rerun only the failed examples and manually validate each passing example's log against its source. ## Quick Start Use the examples-auto-run skill to start the repo examples in auto mode, then report status and summarize any failures from the logs.

Frequently Asked Questions about examples-auto-run

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run all repo examples in auto mode?

Run scripts/run.sh start to execute examples/run_examples.py via uv with auto-input and auto-approve enabled. Optional extras like litellm, sqlalchemy, redis, and temporal are included by default, and logs are written under .tmp/examples-start-logs/.

How do I rerun only the failed examples?

Run scripts/run.sh collect to parse the latest main log into .tmp/examples-rerun.txt, then run scripts/run.sh rerun to execute only those failed entries in auto mode. If the rerun file already exists and is non-empty, invoking the script with no arguments reruns by default.

Can I run the examples in the background?

Yes, pass --background to the start command to detach the process with a pidfile at .tmp/examples-auto-run.pid. Use the status, tail, and stop commands to monitor or terminate the background run.

How do I disable optional dependency extras when running examples?

Set EXAMPLES_UV_EXTRAS to an empty string before starting the run to disable all extras. You can also set it to a custom space-separated list to include only specific extras such as sqlalchemy or redis.

Why do example runs fail inside the Codex sandbox?

Many examples start nested sandboxes, browsers, or service processes that the sandbox blocks, producing errors like sandbox_apply Operation not permitted or npm cache permission errors. Run start or rerun outside the sandbox with escalated permissions unless doing a narrow dry-run.

Does the runner validate that examples behaved correctly?

No, the runner only reports exit codes. After each run, you must manually read each passing example's source and its per-example log to confirm the intended actions and outputs actually occurred.