Conversation
…ependencies from go.mod.
- Add internal/version package - Update chat command to display dynamic version - Update GitHub Action to inject version via ldflags
- Add workspace sandbox to file tools (read_file, write_file, list_dir) with path traversal prevention via validatePath() - Enforce restrict_to_workspace in exec tool; fix double cmd construction - Enhance dangerous command detection with regex patterns covering sudo, separated flags, and more variants - Change config file permissions to 0600 for API key protection - Add config Validate() with range checks for temperature, tokens, etc. - Fix silently ignored os.UserHomeDir() errors in config - Extract shared SplitThink to internal/render package, eliminating duplication between chat.go and telegram.go - Hoist regex compilations to package-level vars in telegram.go - Remove unused toolStyle field and unused port CLI flag - Add 20+ new tests: filesystem path traversal, session persistence, dangerous command variants, multi-turn tool calling, max iterations, render SplitThink coverage - Add CLAUDE.md and development roadmap Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add cron scheduling system (at/every/cron) with persistent JSON storage - Add manage_cron agent tool for AI-driven job management - Add skills loader and installer with GitHub integration - Add CLI commands: channels list/status, cron CRUD, skills CRUD - Enhance status command with gateway, cron, and skills info - Inject skills summary into agent system prompt - Integrate cron service lifecycle into server mode - Update README docs (EN/CN) with new features and CLI reference
Store.Put() now stores a deep copy of the job, and All()/Get() return deep copies. Save() serializes under the read lock. This ensures the tick goroutine's mutations never race with concurrent JSON encoding.
… sharper copy Amp-Thread-ID: https://ampcode.com/threads/T-019c5788-9479-745d-a268-0e79e359c2a4 Co-authored-by: Amp <amp@ampcode.com>
Strip tree/<branch>/ and blob/<branch>/ segments from GitHub web URLs and auto-append SKILL.md when the path points to a directory.
- Agent loop now captures LLM content on every iteration, not only when tool calls are absent, preventing the "Processing complete." fallback from masking actual responses. - Channel manager filters outbound messages with metadata type "heartbeat" so internal health-check logs no longer reach end users.
Prints version, OS, and architecture (e.g. "golem v0.5.3 linux/amd64"). Version is injected at build time via -ldflags.
Use runtime/debug.ReadBuildInfo() so that binaries installed via go install display the tagged module version instead of "dev".
Introduce internal/command/ package with a registry-based slash command framework. Commands are intercepted in processMessage() before the LLM call, working across all channels (TUI, Telegram, Discord, etc.). Built-in commands: /help, /new, /version, /status, /cron, /skills, /memory
Rewrite all command outputs to use proper markdown (bold headers, bullet lists, code backticks) so glamour and Telegram renderers display them correctly. Also auto-generate release notes summary instead of static placeholder.
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.
No description provided.