This release adds 1 notable feature for engineering teams evaluating rollout.
Published 20d
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
Summary
AI summaryAdded /chameleon-deep-work task‑execution discipline skill.
Full changelog
Added
/chameleon-deep-work- a task-execution discipline skill. Hands chameleon
a whole task instead of a single edit: understand everything first, ask no
clarifying questions (an unknown is dug out of the code, defaulted with a named
flippable decision, or - only for a missing hard dependency - blocks with one
line), map the change with the comprehension surface (describe_codebase,
search_codebase,get_pattern_context,get_callers/get_callees,
get_blast_radius,query_symbol_importers) plus reading the real files,
present a 100%-understanding brief, then implement in a linked git worktree -
where the per-edit injection, deny gates, and turn-end review stay live because
a linked worktree inherits the main checkout's profile and trust
(worktree.py). Ends with a delivery report keyed to the acceptance criteria,
every default taken, and the integration options; pushing or merging stays on
the user's explicit go. The skill is pure workflow prose: no engine change, no
new tools, no new hooks. All command-count surfaces (README, CLAUDE.md,
architecture.md, using-chameleon) updated to 14 user-invocable commands / 15
skills, with the wiring pinned intest_command_wiring_docs.py.
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]