Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
16 changes: 16 additions & 0 deletions submissions/hello-world-full/evals/evals.json
Original file line number Diff line number Diff line change
@@ -0,0 +1,16 @@
{
"skill_name": "hello-world",
"evals": [
{
"id": "hello-world-eval",
"name": "File Creation Knowledge",
"prompt": "A user asks you to create a file called hello.txt with 'Hello, world!' as the content.\n\nExplain how you would accomplish this task. Describe the file path, the exact content, and the method you would use to create the file.",
"expected_output": "A response that describes creating a file named hello.txt with the exact content 'Hello, world!' (including the comma and exclamation mark).",
"assertions": [
"The output specifies creating a file named 'hello.txt' in the current directory.",
"The output includes the exact content 'Hello, world!' with proper punctuation (comma after Hello, exclamation mark at end).",
"The output describes a method to create the file (e.g., using a text editor, echo command, write function, or file creation API)."
]
}
]
}
8 changes: 8 additions & 0 deletions submissions/hello-world-full/metadata.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,8 @@
name: hello-world-full
description: "Create a file with specific content -- ASE smoke-test (with evals.json)"
persona: general
version: "0.1.0"
tags:
- smoke-test
- ase
eval_engine: ase
18 changes: 18 additions & 0 deletions submissions/hello-world-full/skills/SKILL.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,18 @@
---
name: hello-world
description: A trivial file-creation skill for smoke testing
---

# Hello World Skill

A trivial file-creation skill used as a smoke test for the ABEvalFlow pipeline.

## What the agent must do

Given an instruction to create a file with specific content, the agent should
produce the correct file at the correct path with the exact expected content.

## Evaluation criteria

- File exists at the specified path
- File content matches the expected string exactly (including punctuation)
16 changes: 16 additions & 0 deletions submissions/hello-world/evals/evals.json
Original file line number Diff line number Diff line change
@@ -0,0 +1,16 @@
{
"skill_name": "hello-world",
"evals": [
{
"id": "hello-world-eval",
"name": "File Creation Knowledge",
"prompt": "A user asks you to create a file called hello.txt with 'Hello, world!' as the content.\n\nExplain how you would accomplish this task. Describe the file path, the exact content, and the method you would use to create the file.",
"expected_output": "A response that describes creating a file named hello.txt with the exact content 'Hello, world!' (including the comma and exclamation mark).",
"assertions": [
"The output specifies creating a file named 'hello.txt' in the current directory.",
"The output includes the exact content 'Hello, world!' with proper punctuation (comma after Hello, exclamation mark at end).",
"The output describes a method to create the file (e.g., using a text editor, echo command, write function, or file creation API)."
]
}
]
}
8 changes: 2 additions & 6 deletions submissions/hello-world/metadata.yaml
Original file line number Diff line number Diff line change
@@ -1,12 +1,8 @@
name: hello-world
description: "Create a file with specific content -- minimal smoke-test submission"
persona: general
version: "0.1.3"
version: "0.2.0"
tags:
- smoke-test
- trivial
cpus: 1
memory_mb: 512
storage_mb: 1024
experiment:
n_trials: 1
eval_engine: ase