Fetching metadata from the HF Docker repository... 
Commit message:
d90c6b8 - cli Add Click-based CLI with memory management commands
- integrations feat: Add AWS Strands Agents SDK integration
- test_cache Introducing headroom wrap
- test_cli fix(mcp): fix uninstall symmetry, move subprocess import, add cli-path tests
- test_compression Fix CI: guard starlette imports, asyncio.run(), deprecate datetime.utcnow()
- test_evals fix(tests): skip HTML extractor tests when trafilatura not installed
- test_integrations Fix LangChain tool_call argument handling for varied message formats
- test_learn fix(learn): pass explicit model in tests to avoid API key requirement
- test_memory feat(router): adaptive compression with Read lifecycle and context-pressure scaling
- test_providers Fix mypy type errors in LiteLLM and OpenAI providers
- test_storage Fix all ruff lint and format errors for CI
- test_transforms Add Kompress: ModernBERT token compressor replacing LLMLingua-2
- 30 Bytes Initial commit: Headroom SDK - LLM context optimization toolkit
- 7.08 kB Add global httpx.ReadTimeout handler for all tests
- 22.4 kB Fix all ruff lint and format errors for CI
- 19.6 kB Add pluggable adapter hooks for CCR, Storage, and TOIN backends
- 12.8 kB Commit message:
- 19.7 kB Fix all ruff lint and format errors for CI
- 41.9 kB fix: improve error handling and add comprehensive test coverage
- 22.6 kB v0.2.2: Add CCR Response Handler, Context Tracker, and restructure docs
- 12.5 kB Fix all ruff lint and format errors for CI
- 21.9 kB Fix security vulnerabilities in memory and CCR systems
- 17.1 kB Fix security vulnerabilities in memory and CCR systems
- 9.17 kB Fix proxy backend bugs: env vars, tool forwarding, and provider support
- 8.74 kB Fix CI: guard starlette imports, asyncio.run(), deprecate datetime.utcnow()
- 44.8 kB fix: improve error handling and add comprehensive test coverage
- 6.96 kB Add compression summaries, multi-provider headers, Dockerfile fix
- 8.42 kB Add compression summaries, multi-provider headers, Dockerfile fix
- 8.99 kB Add compression summaries, multi-provider headers, Dockerfile fix
- 7.81 kB Add Query Echo: re-inject user question after compressed tool outputs
- 10.8 kB Add compression summaries, multi-provider headers, Dockerfile fix
- 19.6 kB fix: dashboard metrics, TTFB tracking, eager LLMLingua loading, and multi-provider consistency
- 16.9 kB Add Query Echo: re-inject user question after compressed tool outputs
- 49.2 kB Update test for lowered TOIN confidence threshold default
- 15.8 kB Fix all ruff lint and format errors for CI
- 23.2 kB Fix Google Gemini conversion to preserve non-text content (Gap #4)
- 12.9 kB Add E2E tests for Google multimodal content preservation
- 8.9 kB Add bounded memory support to HNSWVectorIndex with LRU eviction
- 5.14 kB Add Compression Hooks — extension points for SaaS and advanced customization
- 33.8 kB Slim core dependencies: 2.5GB → 195MB install size
- 25.6 kB fix: improve error handling and add comprehensive test coverage
- 18.4 kB Add Memory Bridge: bidirectional sync between markdown files and Headroom memory
- 9.74 kB Add hierarchical memory system with graph + vector storage
- 24.9 kB Add centralized ML model configuration
- 65.5 kB Fix CI: guard starlette imports, asyncio.run(), deprecate datetime.utcnow()
- 12.2 kB Add memory observability system (Phase 1)
- 15.8 kB Add SQLite vector backend as default and fix HNSW test skipping
- 25.5 kB Add SQLite vector backend as default and fix HNSW test skipping
- 9.41 kB Fix pricing lookup for retired Claude models
- 10.3 kB Commit message:
