Training Log Analyzer
CheckedAnalyzes training logs to identify patterns and insights for improved performance optimization and targeted skill development.
Install on your platform
Showing Claude Code. One command installs this skill into any of the agents below.
- 1
Run this in your project
terminal$ npx skills add onewave-ai/claude-skills --skill training-log-analyzer -a claude-code Uses the open-source
skillsCLI and needs Node.js. Add-gto install it for your user instead of this project.
What we know about this skill
Publisher
onewave-ai
Taken from the source listing. We haven't verified who controls this account.
Install target
Live repository
Not pinned. Installing fetches whatever the repository holds at the time you run the command.
Automated safety checks
This skill passed all 6 checks. We read its files at one commit and looked for the things that make a skill dangerous to run — not for whether it works.
Declares what it does: passed
SKILL.md parses and declares a name and a description.
SKILL.md exists, its frontmatter parses, and it declares a name and a description.
No undeclared network calls: passed
Makes no network calls.
Every host the skill dials — from its scripts, its config endpoints or the commands it tells the agent to run — is either well-known package and API infrastructure or declared in its own frontmatter. Links in documentation are not counted.
No obfuscated payloads: passed
No encoded or obfuscated payloads.
No base64, hex or compressed blob is decoded and then executed, and no encoded shell commands.
No credential access: passed
Reads no keys, tokens, keychains or credential files.
Neither the code nor the SKILL.md instructions read SSH keys, cloud credentials, keychains, token stores, browser cookie databases or dotfiles that hold secrets.
No pipe-to-shell installers: passed
No pipe-to-shell or download-and-run installers.
No `curl … | sh` pattern, in the code or in the instructions, that runs code fetched at run time.
Pinned to a commit: passed
Scanned onewave-ai/claude-skills at f317e08649.
The exact upstream commit SHA that was scanned is recorded, so the result refers to specific bytes.
- Commit
- f317e08649
- Manifest
- training-log-analyzer/SKILL.md
- Scanned
- 1 file, 58 lines
About this skill
The training-log-analyzer skill processes raw AI agent training logs to extract performance metrics, error patterns, and learning trajectories. It transforms unstructured log data into actionable insights for optimizing model behavior during development cycles.
When to use it
- Debugging convergence issues in reinforcement learning or supervised fine-tuning experiments.
- Identifying specific failure modes or hallucination spikes within large-scale training datasets.
- Automating the review of agent interactions to validate alignment with safety guidelines.
- Comparing performance deltas between different hyperparameter configurations or prompt strategies.
Key capabilities
- Parses structured and semi-structured text logs from various training frameworks.
- Aggregates statistical summaries such as loss curves, reward scores, and token usage.
- Flags anomalies and outliers that indicate model instability or data contamination.
- Generates concise reports summarizing the health of a specific training run.
Example prompts
- "Analyze the attached training logs from my last RLHF session and summarize the convergence rate."
- "Identify any instances where the agent's reward score dropped below 0.5 in the provided log file."
- "Extract the top three error patterns causing the model to fail during the validation phase."
Tips & gotchas
Ensure your training logs are formatted consistently or include clear delimiters, as the analyzer relies on structured parsing to function accurately. This tool is designed for post-training analysis; it cannot modify active training runs in real-time.