### CHANGES
- Update toMessages to handle multi-content messages with text and attachments
- Add contentBlocksFromMessage to convert message parts to Anthropic blocks
- Implement support for image URLs including data URLs and base64 images
- Add PDF attachment handling via data URLs and URL-based PDFs
- Introduce parseDataURL for extracting MIME type and data from data URLs
- Create normalizeImageMimeType to standardize supported image MIME types
- Add isPDFURL to detect PDF files from URL paths
- Refactor system content accumulation to use text extraction from parts
- Update tests to verify PDF attachment processing in multi-content messages
## CHANGES
- Add new pattern extending `extract_wisdom` with speaker attribution
- Create README and system.md for the new pattern
- Update pattern_explanations.md with new pattern entry
- Add pattern to suggest_pattern category lists
- Update pattern_descriptions.json with metadata and tags
- Update pattern_extracts.json with pattern content
Adds a pattern that suggests Gas Town (gt) CLI commands based on
natural language descriptions. Covers 85+ commands across work
management, agents, communication, services, diagnostics, and recovery.
Pipe-friendly output: first line is the raw command, enabling:
echo "message the mayor" | fabric -p suggest_gt_command | head -1
Gas Town: https://github.com/steveyegge/gastown
Generated with [Claude Code](https://claude.ai/code)
via [Happy](https://happy.engineering)
Co-Authored-By: Claude <noreply@anthropic.com>
Co-Authored-By: Happy <yesreply@happy.engineering>
## CHANGES
- Add Spotify plugin with OAuth token handling and metadata
- Wire --spotify flag into CLI processing and output
- Register Spotify in plugin setup, env, and registry
- Update shell completions to include --spotify option
- Add i18n strings for Spotify configuration errors
- Add unit and integration tests for Spotify API
- Set gopls integration build tags for workspace
## CHANGES
- Add warning to stderr when using incompatible models with image generation
- Add GPT-5, GPT-5-nano, and GPT-5.2 to supported image generation models
- Create `checkImageGenerationCompatibility` function in OpenAI plugin
- Add comprehensive tests for image generation compatibility warnings
- Add integration test scenarios for CLI image generation workflows
- Suggest gpt-4o as alternative in incompatibility warning messages
- Replace go-git status API with native `git status --porcelain` command
- Fix worktree detection issues caused by go-git library bugs
- Simplify `IsWorkingDirectoryClean` to use CLI output parsing
- Simplify `GetStatusDetails` to return raw porcelain output
- Use native `git rev-parse HEAD` to get commit hash after commit
- Remove unused `os` and `filepath` imports from walker.go
- Remove complex worktree file existence checking logic
Adds test coverage for the Infermatic AI provider in
TestCreateClient to verify the provider exists and
creates a valid client.
Part of #1033: Add Infermatic AI provider support
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
### CHANGES
- Remove value from fractional part error message
- Update overflow check to use float64 for consistency
- Ensure error messages omit unnecessary details for clarity
## CHANGES
- Add NaN and Infinity validation for float64 values
- Add NaN and Infinity validation for float32 values
- Add negative value check for int64 type
- Add negative value check for json.Number type
- Add comprehensive test cases for special float values
- Add test cases for negative int64 and json.Number inputs
- Update line reference comments for validation checks
- Add comprehensive unit tests for `parseOllamaNumCtx` function
- Remove redundant negative value checks in float parsing
- Simplify error messages to avoid exposing internal type info
- Streamline error response in `ollamaChat` handler
- Add helper functions for string containment in tests
- Cover edge cases including overflow, invalid types, and boundaries
This commit fixes the Ollama server /api/chat endpoint which was ignoring
the client-provided num_ctx parameter and global DEFAULT_MODEL_CONTEXT_LENGTH,
always using a hardcoded value of 2048 tokens.
- Add parseOllamaNumCtx() function in ollama.go with type-safe extraction
supporting 6 numeric types (float64, float32, int, int64, json.Number, string)
- Extract num_ctx from client request options in ollamaChat()
- Add ModelContextLength field to ChatRequest struct in chat.go
- Replace hardcoded 2048 with request.ModelContextLength in GetChatter() call
- Platform-aware integer overflow protection for 32-bit systems
- DoS protection via 1,000,000 token maximum limit
- Long string truncation in error messages (50 char limit)
- Sanitized error messages (no internal stdlib details exposed)
- Missing/null num_ctx returns (0, nil) to trigger existing default fallback
- Zero API contract changes
- Invalid values return 400 Bad Request with clear error messages
- All existing tests pass
- Compilation successful with no errors or warnings
Fixes#1942
- Add greybeard_secure_prompt_engineer pattern to create secure system prompts
- Update pattern explanations and renumber existing entries
- Refactor build process to use npm hooks for copying JSON files
- Remove manual web static file copying from extract script
- Update .gitignore to exclude generated data and tmp directories
- Modify suggest_pattern categories to include new security pattern
- Delete redundant web static data file, rely on build hooks
## CHANGES
- Add `Variables` field to `OllamaRequestBody` struct for direct variable passing
- Change `Options` field from empty struct to flexible `map[string]any` type
- Extract variables from top-level `Variables` field or nested `Options.variables`
- Support parsing variables as JSON string or map format
- Pass extracted variables to `PromptRequest` for single message chats
- Pass extracted variables to `PromptRequest` for multi-message chats
- Add `omitempty` JSON tags to optional fields
## CHANGES
- Use int64 for prompt and eval count fields
- Skip sending secondary error message on stream write failure
- Allow non-http schemes and validate host only for address
- Flush response only when writer implements http.Flusher
## CHANGES
- Use Gin request context for outbound HTTP calls
- Send final stream chunk when response write fails
- Capture timing duration once for consistent metrics
- Build final Ollama response via shared helper function
- Validate Fabric base URL scheme is http/https only
- Add clarifying documentation comments for URL and writers
## CHANGES
- Use int64 for `load_duration` JSON field values
- Use int64 for `prompt_eval_duration` JSON field values
- Remove lossy int casts when assigning nanosecond durations
- Keep duration payloads consistent with `total_duration` precision
- Prevent potential overflow on long-running request timing
