dgf1988/codewhale
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

test(fleet): CI-safe multi-worker dogfood smoke with injected failure (#3166)

Browse Source 
Adds a verifiable dogfood smoke that drives several concurrent exec-style
workers (three healthy + one injected-failure that emits an error event and
exits non-zero) through the real host adapter, asserting distinct terminal
pass/fail outcomes — no external services, no model calls, no codewhale binary.
Documents the automated CI smoke vs the manual `codewhale fleet run` path in
the dogfood spec, and is honest that the manager run-loop cutover to drive real
FleetExecutor workers is still in progress.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This commit is contained in:
Hunter B
2026-06-13 01:30:06 -07:00
parent 22980b8788
commit 5106ecfb2d
2 changed files with 76 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/examples/fleet-dogfood.toml
+14 -3
View File
@@ -1,16 +1,27 @@
# Agent Fleet dogfood smoke spec (#3166)
#
# This spec exercises the fleet end-to-end: create a run with two local
# workers, run a lint task and a review task, verify the ledger records
# receipts, and confirm the status surfaces work.
# workers, run a workspace-check task and a protocol-review task, verify the
# ledger records receipts, and confirm the status surfaces work. Each worker is
# a headless `codewhale exec` run (see docs/AGENT_RUNTIME.md).
#
# Run:
# Automated CI-safe smoke (no external services, no model calls):
# cargo test -p codewhale-tui --bins fleet::executor
# It drives several concurrent exec-style workers (with one injected failure)
# through the real host adapter and asserts terminal pass/fail outcomes.
#
# Manual run (drives real `codewhale exec` workers; needs provider creds):
# codewhale fleet run docs/examples/fleet-dogfood.toml --max-workers 2 --once
#
# Then check:
# codewhale fleet status
# codewhale fleet inspect <worker-id-from-status>
# codewhale fleet logs <worker-id-from-status>
#
# NOTE: wiring the manager run loop to drive FleetExecutor for real workers is
# the in-progress cutover; until then the manual run path uses the local
# simulation harness. The automated smoke above already proves the real
# exec-subprocess -> ledger-event path.
name = "dogfood smoke"
labels = { milestone = "v0.8.60", class = "smoke" }