Fund: Live hledger data integration
Broadcom (AVGO) 10-K deep dive
Research: SRAM companies 10-K deep dives — post-Groq/NVDA inference thesis
Research: Data center cooling companies 10-K deep dives
Research: Arista Networks (ANET) 10-K deep dive
Newsreader: remove card border/background from file reader view
Newsreader: list items not indented in document renderer
Newsreader: breadcrumbs reset when viewing file content
Newsreader: Slow Feed Digest view
Adopt Showboat proof-of-work pattern in Coder agent skill
PersonalBrowser skill: install and configure Rodney for personal browser control
Newsreader: auto-fetch full article text for truncated RSS entries
Support ~/work as a second Pipeline domain
Fix agentd buffet provider support (claude-code, codex)
Pipeline smoke test: add module doc to Omni/Pipeline/Git.hs
Task system event/notification layer
fund command - domain budget management
DevBrowser fails without local Playwright browser binaries
Set up Samba file share on beryllium over Tailscale
Benchmark memory system against LoCoMo eval
Interactive step-through debugger
Socket message protocol and handler
Ava dual-channel listener (Telegram + socket)
Pi extension for agent completion notifications
Add user_profiles table for per-user preferences
Pre-compaction memory flush
Graph: IR foundation with Applicative parallelism
Graph: Terminal UI for navigation and composition
Graph: TUI event handling
Graph: TUI rendering
Graph: TUI state and navigation
Agent Graph (epic)
Interactive agent mode (epic)
Research: Hydration Source Provider abstraction
B bild does not propagate langdeps from dep-commented modules through hsGraph
Enforce PromptIR token budgets in Sequential
Wire AdaptiveContextConfig thresholds
Update memory embeddings when merging duplicates
Fix token accounting in semantic/knowledge sections
Review memory improvements and prompt IR
Implement hash-based CI test cache invalidation
Build generic chat adapter layer for multi-platform agent support
Investigate: compare task web vs Studio tasks, plan improvements
Research: auto-detection for best-of-N parallel execution
Integrate weeder for Haskell dead code detection
Migrate from Tailscale to Headscale
Prototype: LLM as universal semantic interpreter (temperature=0)
Design: leverage pure cognitive compute in Bayesian prompt framework
Research: rate-distortion theory for prompt compression
Research: Bayesian calculus for prompt composition
Research: static prompt analysis using open model internals
Implement prompt composition with conventions
Implement empirical prompt compression
Implement heuristic prompt factorization for parallel execution
Add capability-based permissions to agentd workflows
PIL Daily Growth Operations
Investigate git-branchless test worktree crash on push
Analytics and funnel tracking
Drip scheduler for follow-up campaigns
Execution details overwritten on retries - should accumulate activity records