monitor-experiment
A utility for tracking Beaker experiment status and automatically retrieving logs upon completion.
Install
mkdir -p .claude/skills/monitor-experiment && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/2679" && unzip -o skill.zip -d .claude/skills/monitor-experiment && rm skill.zipInstalls to .claude/skills/monitor-experiment
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Monitor Beaker experiments until completion. Use when the user asks to monitor, check, or track a Beaker experiment.Key capabilities
- →Poll Beaker experiment status
- →Detect experiment completion
- →Report success or failure status
- →Fetch logs for failed experiments
- →Stream logs in real-time
How it works
The skill periodically polls the Beaker API for an experiment's status until it exits, then conditionally retrieves logs if the exit code is non-zero.
Inputs & outputs
When to use monitor-experiment
- →Monitor experiment progress
- →Check Beaker job status
- →Retrieve logs for failed experiments
- →Automate experiment tracking
About this skill
Monitor Beaker Experiment
Instructions
When monitoring a Beaker experiment:
- Status is per job, and a retried experiment holds several jobs. For each task and replica rank, read the most recently created job, never
jobs[0]:beaker experiment get <experiment-id> --format json | jq -c '.[0].jobs | group_by([.execution.task, .execution.replicaRank]) | map(max_by(.status.created) | {id, rank: .execution.replicaRank, status: (.status | {finalized, exitCode, canceledFor})})' - A job is done once
status.finalizedis set. A cancelled job has noexitCode: readcanceledFor. Beaker can replace a cancelled job (a healthcheck swap, or a preemption when the spec setsautoResume), so re-run step 1 a few minutes later before calling a cancellation final: a newer attempt for that rank means the experiment is still running.exitCode0 is success; otherwise show the failing rank'sbeaker job logs <job-id> --tail 200. - If a job is not done, check again in about a minute. Claude Code blocks a foreground
sleep: use Monitor or a background loop. Codex cansleep. - Continue until every job is done or the user asks you to stop.
Stream a running job's logs with beaker job logs --follow <job-id>, using a job id from step 1.
When not to use it
- →Monitoring non-Beaker jobs
- →Local process tracking
Prerequisites
Limitations
- →Only supports Beaker experiments
- →Requires active network connection to Beaker
How it compares
It replaces manual polling and log retrieval with an automated loop that handles status checking and error reporting.
Compared to similar skills
monitor-experiment side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| monitor-experiment (this skill) | 2 | 10mo | Review | Beginner |
| aggregating-crypto-news | 4 | 4mo | Review | Beginner |
| status | 0 | 9mo | No flags | Beginner |
| news-fetch | 0 | 10mo | No flags | Intermediate |
Try saying
Example prompts that trigger this skill in your AI assistant.
More by allenai
View all by allenai →You might also like
aggregating-crypto-news
jeremylongshore
Aggregate breaking cryptocurrency news from 50+ sources including CoinDesk, CoinTelegraph, The Block, and Decrypt. Use when needing to monitor crypto market news, track announcements, or find coin-specific updates. Trigger with phrases like "get crypto news", "latest Bitcoin headlines", "DeFi announcements", "scan for breaking news", or "check crypto updates".
status
Defiect
Check the progress of the current or most recent deep research run. Shows stage, source counts, claim counts, and any issues.
news-fetch
Krual-T
Search and archive significant global events (7-day window). Topics: international conflicts, trade/China policy, and AI/Agent updates. Include entity tracking, semantic dedupe against news.jsonl, and atomic archival via recorder.
cto-engineering-metrics
rinaldofesta
Expert methodology for defining, tracking, and interpreting engineering performance metrics including DORA, team health, productivity, and executive reporting.
literature-review
K-Dense-AI
Conduct comprehensive, systematic literature reviews using multiple academic databases (PubMed, arXiv, bioRxiv, Semantic Scholar, etc.). This skill should be used when conducting systematic literature reviews, meta-analyses, research synthesis, or comprehensive literature searches across biomedical, scientific, and technical domains. Creates professionally formatted markdown documents and PDFs with verified citations in multiple citation styles (APA, Nature, Vancouver, etc.).
openalex-database
davila7
Query and analyze scholarly literature using the OpenAlex database. This skill should be used when searching for academic papers, analyzing research trends, finding works by authors or institutions, tracking citations, discovering open access publications, or conducting bibliometric analysis across 240M+ scholarly works. Use for literature searches, research output analysis, citation analysis, and academic database queries.