Ready Queue (96 tasks)

t-822OpenP2

Refactor HydrationConfig to stacked HydrationSource list

t-821OpenP2

Wire knowledge graph traversal into passive hydration

t-820OpenP2

Wire knowledge graph traversal into passive hydration

t-814OpenP2

Prevent persistent agent 'promise loop' (says it will work but does not execute tools)

t-813OpenP2

Make omni .envrc idempotent/non-noisy for non-interactive direnv exec

t-810OpenP2

agentd top: j/k navigation, ^C watch exit, Activity cache refresh

t-807OpenP2

Refactor shared systemd helpers across agentd and deployer

t-795OpenP2

Fix typecheck.sh jq parsing for LocalRun rundeps

t-791OpenP2

sd: investigate low acceptance rate + verify idle A100s shut down

t-790In ProgressP2

Intent: Impl-vs-spec verification

t-785OpenP2

Persistent agent env drift can set AGENTD_AGENT_COMMAND=agent and break services

t-782OpenP2

Fix ambiguous Data.ByteString.Base64 import in Omni/Health/Intervals

t-780OpenP2

Design and implement mkSkill Nix helper + skill CLI

t-779OpenP2

Align Anthropic OAuth flow with pi-mono

t-771OpenP2

Notify owner on LLM provider failover

t-765OpenP2

workflows: add system_prompt to reviewer + integrator frontmatter; drop obsolete coder/overseer

t-762OpenP1

Pipeline scheduler: verify commits reach live before marking status=done (fixes orphan-commit bug)

t-761OpenP2

Port pi's system prompt to Omni/Agent coding agents

t-760OpenP2

Design compaction strategy for Omni/Agent coding agent

t-758OpenP2

Fix Omni/Ipt.py build failure

t-755OpenP2

Research: unify trace compatibility for persistent sessions

t-752OpenP3

Skill drift: ui-preview example references removed agentd-studio path

t-750In ProgressP2

Cleanup: remove legacy in-process agent code from Omni.Agentd.Daemon

t-749OpenP2

Bug: task CLI fails to read mixed-case enum values in tasks.db

t-748OpenP2

Bug: agentd service restart can hang in stop-sigterm for ~2m

t-738OpenP2

Intent Phase 4 follow-up: fix exec_cmd code type (Nat→Int)

t-737OpenP2

Intent Phase 4 follow-up: align :foreign ordering semantics (parser/docs/tests)

t-736OpenP2

Intent Phase 4 follow-up: include declForeign in hash/store persistence

t-735OpenP2

Intent Phase 4 follow-up: make :foreign python executable in codegen

t-734OpenP2

Intent: include regression cases in default test workflow

t-733OpenP2

Intent: resolve remaining LANG.md contract contradictions

t-732OpenP2

Intent: make query-holds KB scoping strict by default

t-731OpenP2

Phase 4: Health metrics at /ava/health

t-730OpenP2

Phase 3: Engagement audit at /ava/engagement

t-729OpenP2

Phase 3: Context inspector

t-728OpenP2

Phase 2: Memory browser at /ava/memories

t-727OpenP2

Phase 2: Message detail view at /ava/chats/:chatId/messages/:msgId

t-724OpenP2

Ava Audit Dashboard

t-717OpenP2

Quant Phase 2: Alpha + Portfolio + Web

t-708OpenP2

Quant Phase 1: Data + Signals + Eval

t-702OpenP2

Secrets not provisioned after reboot

t-700OpenP2

Web architecture improvements (epic)

t-699OpenP2

Set up SMTP secrets for fund-digest service

t-696OpenP2

Migrate fund queries from hledger CLI to hledger-lib Haskell API

t-695OpenP2

Agentd daemon pi.log writing is broken - handle closed immediately

t-694.7OpenP2

Send to Review: integrate agent work into live branch

t-694.6OpenP2

Worktree diff view and file browser

t-690OpenP2

Bug: ship/push may report manifest update without persisting

t-685OpenP2

Add torch, mamba-ssm, causal-conv1d to omnirepo Python nix deps

t-684OpenP2

Guard against web deploy rollback from divergent worktrees

t-683OpenP2

MVP 3: Diverse model review (devil's advocate)

t-682OpenP2

MVP 2: Single-model auto-spec refinement loop

t-681OpenP2

MVP 1: Executor GO/NO-GO gate

t-680OpenP2

MVP 0: Baseline metrics for task pipeline

t-679OpenP2

Epic: Automated Task Up-Speccing Pipeline

t-677OpenP2

JIT context rehydration via request_context tool

t-675.1OpenP1

MVP 0: Baseline metrics instrumentation

t-667OpenP2

General OAuth 2.0 integration for Telegram bot

t-666OpenP2

Agent auth path drift: pipeline/agentd uses ~/.local/share/agent but fresh creds are in ~/.pi/agent

t-665OpenP2

Agentd: avoid zombie 'running' status when container exits before terminal event

t-664OpenP2

Pipeline: don't move failed runs to Review

t-658OpenP2

Investigate mbsync duplicate UID errors in Maildir

t-655OpenP2

Task system: validate namespace exists in repo when setting it

t-653OpenP2

Broadcom (AVGO) 10-K deep dive

t-652OpenP2

Research: SRAM companies 10-K deep dives — post-Groq/NVDA inference thesis

t-651OpenP2

Research: Data center cooling companies 10-K deep dives

t-650OpenP2

Research: Arista Networks (ANET) 10-K deep dive

t-647In ProgressP2

Newsreader: breadcrumbs reset when viewing file content

t-631OpenP2

Support ~/work as a second Pipeline domain

t-630OpenP2

Fix agentd buffet provider support (claude-code, codex)

t-622OpenP2

fund command - domain budget management

t-620OpenP2

DevBrowser fails without local Playwright browser binaries

t-602OpenP2

Benchmark memory system against LoCoMo eval

t-549.3OpenP2

Interactive step-through debugger

t-538.3OpenP2

Socket message protocol and handler

t-538.2OpenP2

Ava dual-channel listener (Telegram + socket)

t-538.1In ProgressP2

Pi extension for agent completion notifications

t-535OpenP2

Add user_profiles table for per-user preferences

t-533OpenP2

Pre-compaction memory flush

t-529OpenP2

Graph: Terminal UI for navigation and composition

t-522OpenP2

Graph: TUI event handling

t-521OpenP2

Graph: TUI rendering

t-520OpenP2

Graph: TUI state and navigation

t-513OpenP2

Agent Graph (epic)

t-486OpenP2

Enforce PromptIR token budgets in Sequential

t-485OpenP2

Wire AdaptiveContextConfig thresholds

t-481OpenP2

Review memory improvements and prompt IR

t-455OpenP3

Research: auto-detection for best-of-N parallel execution

t-446In ProgressP2

Integrate weeder for Haskell dead code detection

t-401OpenP2

Prototype: LLM as universal semantic interpreter (temperature=0)

t-400OpenP2

Design: leverage pure cognitive compute in Bayesian prompt framework

t-399OpenP2

Research: rate-distortion theory for prompt compression

t-398In ProgressP2

Research: Bayesian calculus for prompt composition

t-397OpenP2

Research: static prompt analysis using open model internals

t-362OpenP2

Add capability-based permissions to agentd workflows

t-265.7OpenP3

Drip scheduler for follow-up campaigns