AI agentAn AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.Full definition → SDK agent hookA script the harness runs automatically at a fixed point in an agent's loop — before a tool call, after an edit, on session start.Full definition → callbacks can now fail closed with onFailure: "block", so a broken safety hook no longer lets an action through. The release also adds a cost report API and new OTel attributes for tracking spend.
Key takeaways · AI-distilled
claude -p used to exit 0 when a full disk or quota lost its transcript or ~/.claude.json writes. In 2.1.317 it prints an error and exits 1 in text and json output, though the notes say full NFS exports may still be missed.
OpenTelemetry exports behind a proxy that never answers used to hang indefinitely. They now time out after 10 seconds, adjustable up to 30 seconds with OTEL_EXPORTER_OTLP_TIMEOUT.
The Read tool no longer returns a "file unchanged since your last Read" stub, so rereading an unchanged file now returns its full content.
Subagents and teammates running on a different model from the main conversation now use that model's saved effort level or its default, instead of inheriting the session's effort level.
Claude previously treated partial Grep and Glob results as complete when a search hit its time limit; this release fixes that.
Terms in this piece · Glossary
agent hook — A script the harness runs automatically at a fixed point in an agent's loop — before a tool call, after an edit, on session start.
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
MCP — The Model Context Protocol — an open standard that lets any AI assistant plug into any tool or data source without custom integration code.
