Areno Debug Runtime

What it does

When your agent uses it

• Runtime evidence must identify the first causal stage
• Routine capacity planning without a failure

Example prompts

• “/areno-debug-runtime”

Requirements

• Python 3

Workflow steps

7 steps, taken from the first numbered list in SKILL.md.

1. 1Preserve exact command, commit, config, earliest logs, and worker exit data.
2. 2Classify config/load, prefill, decode, reward, scoring, train forward, backward, optimizer, save, or distributed teardown.
3. 3For multi-rank output, group signatures and prioritize the earliest distinct exception over secondary NCCL watchdog failures.
4. 4Distinguish compilation/autotune work from deadlock using elapsed time and stacks.
5. 5Reproduce with the same semantic workload at the smallest topology that still fails.
6. 6Correct the owning layer. Required kernels must fail loudly; do not introduce silent fallback.
7. 7Re-run the reproduction and the original bounded path.

What it can do on your machine

Read from SKILL.md and the folder at commit ce35e0d. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Ships 3 files in scripts/ (Python), which the agent can run.

  Shell commands in SKILL.md call:

  • python

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Areno Debug Runtime loads about 486 tokens when it runs, and up to ~786 if it reads all its reference files. Until then it costs about 65 tokens; SKILL.md has 184 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~65

When it runs · the whole SKILL.md, loaded when a task matches

~486

With references · SKILL.md plus every file in references/, read only if the agent opens them

~786

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from inclusionAI/AReno at commit ce35e0d, republished under its Apache-2.0 licence (© inclusionAI). 184 words, ~486 tokens.

name

areno-debug-runtime

description

Diagnose failed, hung, slow, OOM, NaN, illegal-memory-access, NCCL, compilation, rollout, or training runs in AReno. Use when runtime evidence must identify the first causal stage. Do not use for routine capacity planning without a failure.

python .agents/skills/areno-debug-runtime/scripts/summarize_traceback.py run.log
python .agents/skills/areno-debug-runtime/scripts/process_snapshot.py --pid <pid>
python .agents/skills/areno-debug-runtime/scripts/inspect_core.py <core-file> --executable <python>

© inclusionAI, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 6 other files (scripts, references) in .agents/skills/areno-debug-runtime of inclusionAI/AReno.

• SKILL.md
• agents/openai.yaml
• references/error-taxonomy.md
• references/nan-triage.md
• scripts/inspect_core.py
• scripts/process_snapshot.py
• scripts/summarize_traceback.py

Open the folder on GitHubat commit ce35e0d