Claude Code v2.1.293 ·
Session safety-stop count now documents what it counts and when it resets
The result field counting safety stops covers refusals, content-filter stops and sometimes safety-monitor blocks, and resets on resume and /clear
Under the hoodTier: how much it should matter to you
1Useful: my rating, 1 to 5
1Signal: worth watching, 1 to 5
AgentsArea: what it touches
Internal ChangesKind: in v2.1.293,
Internal ChangesSection of the release
Unclear The name of this result field is not known.
What
Results include a count of how many times a safety system stopped one of the session's model calls. Its description now says it covers:
- refusals by the model
- content-filter stops
- in some builds, blocks by a safety monitor
The count is never saved, so it starts again from zero when you resume a session or run /clear.
Why
Programs built on Claude Code can use this count to notice refusals, as long as they allow for it resetting.
Something disagreesSomething we can check disagrees with this entry, or the writer said they could not settle it.
The writer flagged doubtThe name of this result field is not known.
Anthropic's release notes agreeFixed a claude.ai-synced skill's edited description sometimes not reaching the model until a new conversation or /clear
See this entry in the whole of v2.1.293 →
