v1.1.38

Changelog

• 1f63c54 feat(oauth): register gpt-5.6-terra + gpt-5.6-luna Codex routes (#769)

v1.1.37

Changelog

• e377e48 feat(oauth): reasoning params (adaptive thinking + effort) + output caps for auth/* lanes (#768)

v1.1.36

Changelog

• 965ca61 feat(llm): compose per-role model via PluginBundle.models + env (#767)

v1.1.34

Changelog

• 2bc249a fix(write_file): per-model max_tokens + actionable "already exists" error (#757)

v1.1.33

Changelog

• 4a8c953 fix(agents): restore cve_lookup on exploit/exploiter/cloud/mobile/detector (#755)

v1.1.29 — litellm auth/* prompt caching

Wires up Anthropic prompt caching on the Claude Code OAuth (auth/*) LiteLLM path — both the cache-token metering and the actual conversation-history caching. Long, tool-heavy runs now re-read the repeated context prefix at 0.1× instead of full price.

What changed

• Meter cache tokens. The OAuth handler (config/claude_code_handler.py) now surfaces cache_creation_input_tokens / cache_read_input_tokens from the Anthropic response in the returned usage (non-streaming and streaming), and includes them in prompt_tokens so LiteLLM's cost path prices cache reads (0.1×) / writes (1.25×) instead of silently dropping them.
• Cache the conversation, not just the system prompt. A cache_control breakpoint is now stamped on the last message block (incremental / auto-advancing), so the whole system + tool-result / history prefix is cached and re-read across turns. System breakpoints capped to 3 to stay within Anthropic's global limit of 4.

Impact

Previously only the system prompt carried cache_control and the response's cache counters were discarded, so caching was invisible/unpriced and the dominant tool-result/history bulk was never cached. This release closes both gaps.

Verification (live, real Anthropic via OAuth)

• System caching: call 1 cache_creation=11427 → call 2 cache_read=11427.
• Multi-turn history caching: Turn A cache_creation=9086 → Turn B cache_read=9086, cache_creation=35 (full prefix served at 0.1×, only the new delta written).

Downstream

Consumers that build FROM ghcr.io/purpleailab/decepticon-litellm:1.1.29 must also add cache pricing (cache_read_input_token_cost / cache_creation_input_token_cost) to any auth/* model_info that overrides the built-in cost map, or cache tokens price at $0.

Commits

• feat(litellm): meter + cache the auth/* OAuth path (prompt caching) (#748)

v1.1.28

Changelog

• a0ee6dd feat(reporting): deterministic CVSS 3.1 scoring tool in core (#746)

v1.1.24

Changelog

• 3067b1d fix(sandbox): widen connection retry to cover per-engagement sandbox boot (#737)

v1.1.23

Changelog

• b2256de fix(filesystem): route filesystem tools to the per-run sandbox (#727)

v1.1.22

Changelog

• 7a5dee1 Merge pull request #725 from PurpleAILAB/feat/proxy-key-state-channel