Feat/image support#21
Merged
Merged
Conversation
- New ImageAttachment type with pure-std base64 encoding, file validation (size/mime), and data: URI generation - /image command: attach, list, drop, clear pending images - Images flow through Message -> providers automatically - Ollama provider: passes images via ollama-rs ChatMessage - OpenAI-compat (llama.cpp, vLLM): multipart content array with image_url parts - screenshot tool: model can request user to capture screen - Session persistence: base64 data stored in JSONL files for self-contained sessions - pending_images held in CommandContext, cleared after send
Prompts were ~240 lines with massive duplication. Cut to 44 lines (~120 tokens) while preserving essential behaviour. header.md 9 lines — identity + 6 core rules agent.md 5 lines — mode role + switch hint planning.md 24 lines — process, format, handoff research.md 3 lines — role + citation rule casual.md 3 lines — mentions web_search/web_fetch The model already knows how to code, plan, and research. The prompt only encodes TinyHarness-specific behaviour: use glob not find, edit before write, 3-strike rule, no hardcoded secrets. Also: Casual mode now has web_search + web_fetch tools (previously had zero tools — completely isolated).
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
feat: image attachment support & leaner system prompts