Blog
News, launch notes, guides, and comparisons for builders exploring reusable AI agent skills, trust signals, and ecosystem shifts.
Latest posts
Follow product updates, editorial pieces, and practical explainers that help users navigate the marketplace and adjacent agent ecosystems.
-
Scan LLM systems for jailbreaks, prompt injections, and unsafe behaviors with garak
Probe a model or agent stack with adversarial test suites so safety failures show up before deployment or review.
-
Score model outputs with reusable evaluator prompts and metrics using autoevals
Apply reusable evaluators to model outputs when you need lightweight scoring, rationale capture, or quick eval loops in code.
-
Turn Markdown runbooks into executable, parameterized procedures with Runme
Run named steps directly from Markdown runbooks so operational docs stay executable instead of drifting into stale reference text.
-
Filter prompts and model outputs for injection, secrets, toxicity, and policy risks with LLM Guard
Screen prompts and responses with input and output scanners before an LLM interaction reaches production users or downstream systems.
-
Generate LLM fine-tuning, RAG, and eval datasets from source material with easy-dataset
Turn raw documents into structured fine-tuning, RAG, and evaluation datasets when the real job is dataset preparation, not generic document parsing.
-
Capture and search AI coding sessions alongside Git history with Entire CLI
Record AI coding sessions next to commits so an agent can recover why code changed, resume from checkpoints, and hand off work cleanly.
-
Sweep GitHub for leaked secrets and exposed credentials with git-hound
Search public GitHub broadly for leaked secrets and triage exposures when the workflow is recon and remediation, not generic secret scanning.
-
Generate conventional-commit release notes for automated ship workflows with semantic-release release-notes-generator
Generate structured release notes from Conventional Commits inside automated release pipelines when changelog generation is the specific job to do.
-
Scan repositories for Shai-Hulud 2.0 supply-chain indicators with the detector action
Check repositories and CI surfaces for Shai-Hulud 2.0 compromise indicators when the task is targeted supply-chain triage, not generic malware scanning.
-
Triage active security incidents with AI-augmented workflows in Valhuntir CLI
Guide live digital-forensics and incident-response work with human approval gates when the job is evidence review and triage, not general MCP setup.