Add executor tool loop (A12/A15) — step 5b #12

Merged

dev-agent merged 3 commits from executor-loop into main 2026-09-07 20:35:33 +00:00

Conversation 1 Commits 3 Files Changed 10

+937 -30

3 Commits

Author	SHA1	Message	Date
dongarandClaude Opus 5	bacf65dbe2	Fix AWAIT_INPUT deadlock and make the live milestone test robust ... Found while chasing a flaky live test, but the flakiness was a real bug: only the ask_human path recorded a pending item, so AWAIT_INPUT reached any other way — planning failure, stopped milestone, boot reconciliation — refused every operator reply with "no pending question to answer" and the agent could never be recovered. The operator had no way back short of wiping state. Every path into AWAIT_INPUT now records what a reply should resume, and answering dispatches on it: planning failures and questions resume planning, a stopped milestone returns to its still-approved gate, and the fallback says what commands work instead of dead-ending. Two regression tests cover the recovery paths. The live test also scripted a fixed approve/approve/start sequence, which a real model breaks by legitimately calling ask_human at a gate. It now plays the operator adaptively, responding to whatever the agent actually did. Five consecutive live runs pass. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>	2026-09-07 16:46:07 +00:00
dongarandClaude Opus 5	999a38a705	Fix process-group kill test to work without an init process ... os.kill(pid, 0) succeeds on a zombie, and a killed orphan stays a zombie until reaped — which never happens inside a CI container with no init. The test read that as "still alive" and failed there while passing locally. It now reads the process state from /proc, with a control assertion so a broken liveness check cannot pass silently. The group kill itself was correct. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>	2026-09-07 16:37:38 +00:00
dongarandClaude Opus 5	0fd0b2e4a5	Add executor tool loop (A12/A15) — step 5b ... - tools.py: four flat tools (read_file, write_file, list_dir, run_command) per A15's reduced manifest. Deliberately unsandboxed: the VM is the trust boundary (B3) and tool use is auto-approved inside an approved unit of work (A4), so path confinement would add failure modes without adding safety. What is enforced is operational — timeout with process-group kill so a backgrounded server cannot outlive its command, output truncation so a build log cannot eat the context window, closed stdin so prompts fail fast, and a scrubbed environment so project code never sees the agent's credentials - executor.py: renders the log, calls the executor role with the reduced manifest, dispatches tool calls, and checkpoints every result (A13). Escalation is observable, never self-assessed (A12): consecutive command failures and a hard turn cap. Failed reads do not count — looking around is not failing - Until delegation exists (step 7), escalation gates to the human with the failures attached - control.py: start milestone now runs the executor; a stopped milestone gates to AWAIT_INPUT Two bugs found by running it live against real models: - A turn-ending gate tool (submit_plan) left a tool_call with no tool_result, which providers reject outright, so every planning turn after the first failed with a 400. The planner now records the gate as the call's outcome, which is also the truthful transcript - The renderer now emits a placeholder for any tool_call lacking a result, so a crash between call and result (A3) cannot poison the transcript permanently Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>	2026-09-07 16:35:39 +00:00