This release fixes issues for SREs watching stability and regressions.
Published 1mo
Developer Productivity
✓ No known CVEs patched
✓ No known CVEs patched in this version
Topics
ai-coding-assistant
archetype
ast-analysis
claude-code
claude-plugin
code-conventions
+12 more
code-review
code-style
developer-tools
javascript
linter
mcp
mcp-server
prompt-engineering
python
ruby
rails
typescript
Affected surfaces
auth
Summary
AI summaryEnforce mode no longer hard‑blocks non‑code files containing "eval(" or credential text.
Full changelog
Fixed
- Enforce mode no longer hard-blocks on
eval(/credential text in non-code
files. The archetype-independenteval-callandsecret-detected-in-content
rules ran on raw content, so a literaleval(or a credential-shaped token in
markdown / plain-text / config PROSE (e.g. documentation that explains the rules,
or a CHANGELOG entry) was treated as a runnable sink. Under the new enforce
default this turn-trapped a session with no escape — a non-code file cannot carry
an inline// chameleon-ignoredirective.eval-callis now gated to recognized
code languages, and a newblock_eligible_on_filegate drops the
archetype-independent rules from the BLOCK set on anydetect_language()-None
file (they remain advisory). Applied at every block/arming site, including the
with-archetype paths reachable by a legacy extension-blindpaths_pattern.
Enforcement on real code (.ts/.js/.rb/.py) is unchanged: a real
eval()/exec()or a committed credential still blocks.
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About Chameleon
All releases →Beta — feedback welcome: [email protected]