hello-aiase

name	hello-aiase
description	Minimal smoke-test skill — echoes a greeting in the AIASE 2026 output contract. Use to verify the Hermes ↔ LiteLLM ↔ skill pipeline is wired up correctly.
version	1.0.0
metadata	{"hermes":{"tags":["smoke-test","aiase-2026"],"category":"utility"}}

Hello AIASE — Smoke Test

When to Use

When the user invokes /hello-aiase with a JSON payload like {"name":"world"}. Use this skill to confirm Hermes can:

1. discover and load this skill,
2. invoke the script under scripts/,
3. relay the resulting fenced JSON block as the final output.

This skill performs no LLM reasoning — it just echoes structured input. If /hello-aiase works end-to-end, the rest of your AIASE 2026 setup is wired correctly.

Procedure

1. Take the entire JSON payload from the user message verbatim (it should be an object; if missing, default to {}).

2. Extract task_id (default to an empty string if missing) and name (default to "world" if missing).

3. Use the terminal tool (do not use process/background tools) with the absolute path to run:

   python3 <skill_dir>/scripts/run.py --task_id "<task_id>" --name "<name>"
   

   where <skill_dir> is the directory Hermes reports when loading this skill.

4. scripts/run.py atomically writes the final result to the result file (path from the environment variable AIASE_RESULT_PATH, falling back to ./aiase_result.json). You do not need to output or repeat the JSON in the chat message.

Pitfalls

• Do not use process/background tools to invoke the script; use the synchronous terminal tool.
• Do not call the LLM to "improve" the greeting — this skill is intentionally deterministic.
• Do not output a fenced JSON block in the final chat message; the grader reads the result file, not the conversation.

Verification

• The result file must exist and be a valid JSON object with fields task_id, greeting, and ok=true.
• If input contains task_id, the output task_id must equal the input task_id.