| #801 |
Stage 3 epic: agent specs + call_agent handoffs (#agentic)
ALTER highway.agents to carry a full run_goal spec (system_prompt, route, tool_catalog, tool_config,...
|
closed |
medium |
2026-07-02 18:35 |
- |
|
| #800 |
Stage 2 epic: agent sessions/threads (memory across run_goal)
Give run_goal cross-call memory via dedicated tables (agent_sessions, session_messages), NOT workflo...
|
closed |
medium |
2026-07-02 12:52 |
- |
|
| #799 |
Structured output: bare ollama format constraint yields empty replies from cloud thinking models
First live structured run: glm-5:cloud under ollama native format returned EMPTY content (not valid ...
|
closed |
medium |
2026-07-01 23:24 |
- |
|
| #798 |
Stage 1 epic: structured output through the LLM gateway + Anthropic prompt caching
Agentic plan Stage 1: (a) response_format JSON-schema param through call_llm to all providers (opena...
|
closed |
high |
2026-07-01 23:01 |
- |
|
| #797 |
send_signal was broken outright: jsonb = json operator error in dedup query
SignalService.send_signal dedup SELECT compares jsonb column signal_payload with a psycopg Json-type...
|
closed |
high |
2026-07-01 22:45 |
- |
|
| #796 |
GET /workflows/<id> returns 500 for paused (and orphaned) runs - response enum missing statuses
spectree response validation: WorkflowStatus enum in api/openapi/models/{workflows,common}.py lacked...
|
closed |
high |
2026-07-01 22:38 |
- |
|
| #795 |
MCP run_goal cannot select a model route
highway-mcp server.py _agent_dsl_for hardcodes route=agent. Add optional route param to run_goal (va...
|
closed |
medium |
2026-07-01 22:22 |
- |
|
| #794 |
reaction_agent hardcodes minimax-m3:cloud instead of the agent route
api/services/reactions.py _REACTION_AGENT_DSL hardcodes provider/model minimax-m3:cloud, bypassing [...
|
closed |
high |
2026-07-01 22:22 |
- |
|
| #793 |
make test targets reference nonexistent test-runner compose service
Makefile test/test-unit/test-integration/test-k run 'docker compose --profile test run --rm test-run...
|
closed |
medium |
2026-07-01 22:22 |
- |
|
| #792 |
Signals never wake a sleeping run (poll-only, asymmetric with events/approvals)
send_signal (engine/services/signal_service.py) inserts into workflow_signals; consumption is poll-b...
|
closed |
high |
2026-07-01 22:22 |
- |
|
| #791 |
Pause is advisory: engine keeps executing tasks of a paused workflow run
POST /workflows/<id>/pause (api/blueprints/v1/workflows.py:~3613) only sets workflow_run.status='pau...
|
closed |
critical |
2026-07-01 22:21 |
- |
|
| #790 |
Stage 0 epic: agentic substrate correctness (pause enforcement, signal wake, route fixes)
Fresh agentic review vs LangGraph/OpenAI SDK/Temporal found substrate-credibility bugs and hygiene i...
|
closed |
high |
2026-07-01 22:21 |
- |
|
| #789 |
Orchestrator capped at per-LEAF token budget (100k) + budget check on resume drops an already-approved action
run_goal (orchestrator) defaults token_budget to max_tokens_per_leaf (100k) when None -- but that's ...
|
closed |
high |
2026-06-30 23:04 |
- |
|
| #788 |
Agent loop can 'complete' at turn 0 with zero tool calls on a structured/constrained goal
A run_goal_deferred with a 'Step 1: delegate... do NOT research yourself; Step 2: do A/B/C' goal com...
|
closed |
high |
2026-06-30 00:48 |
- |
|
| #787 |
HITL gates reads same as side-effects -> approval spam on research (research->approve->act)
run_goal/_deferred with approval gates tools.http.request regardless of whether it's a READ (GET to ...
|
closed |
high |
2026-06-30 00:48 |
- |
|
| #786 |
Switch agent LLM route to ollama/glm-5:cloud (2x faster than qwen3.5:397b)
Benchmarked Ollama Cloud models for the [llm.routes] agent route (controlled: identical prompt, num_...
|
closed |
medium |
2026-06-30 00:08 |
- |
|
| #785 |
schedule_goal crons collide: all share job_name agent_sched_<def_id[:8]> (fixed SCHEDULED_AGENT_DSL)
Same root cause as #784 but for schedule_goal (recurring crons). job_name = 'agent_sched_'+def_id[:8...
|
closed |
high |
2026-06-29 23:19 |
- |
|
| #784 |
watch_gmail poll: stop+recreate leaves watch inactive (durable_cron idempotent on shared job_name)
All poll watches share job_name 'agent_sched_'+def_id[:8] where def_id is the FIXED SCHEDULED_AGENT_...
|
closed |
high |
2026-06-29 22:53 |
- |
|
| #783 |
Gmail auto-responder dedup is non-deterministic (LLM-judged, sender-per-day) -> 26-min delay + skipped/duplicate replies
Watch auto-responder (_watch_goal in highway-mcp/server.py) enforces reply-idempotency via an LLM IN...
|
closed |
high |
2026-06-29 20:19 |
- |
|
| #782 |
Agent eval/regression harness: trace-based assertions (routing, over-fetch, delivery, cost) + selftest
Built the regression harness (highway-mcp/eval_agent.py) the strategy called for. Upgrades the prior...
|
closed |
high |
2026-06-29 18:30 |
- |
|