Files

T

kyleandClaude Fable 5.1 e156975649 Add M2b plan: ten tasks, tests, recordings and the first system prompt

The tasks build the agent loop on M2a's client: channel messages and
the usage record in proto, four config tables, the tool port and
registry with find_tool and call_tool, the baseline and replay, the
session store, the turn loop with its limits and the append-only
property test, the channel server, loopd serve, bxctl chat, and the
device checks including a four-turn conversation with a restart.

Checked against a private reference implementation: the gate passes
after every task in order, the new suites pass under CPU load, and the
reference passes make verify-device on straylight with no cache loss.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

2026-09-18 17:19:02 -07:00

2.2 KiB

Raw Blame History

M2b task 10: verification on the real server

Branch: m2b (run git switch m2b; git status --short must be empty, otherwise stop) Commit subject: Extend on-device verification to the agent loop

Goal

Run the whole loop against straylight: the token budget of the real baseline, and a four-turn conversation through loopd serve and bxctl chat with a loopd restart in the middle. You write no library code. If a check fails, that is a finding. Do not change a test or a limit.

Files

Copy: crates/loopd/tests/device.rs (replaces the M2a one; the four M2a checks are still in it), Makefile (only verify-device changes: it now also passes BOXMAKER_BXCTL), and config/system.md (the first system prompt; the device test reads it from the repository)
Modify: docs/implementer-log.md

Steps

1. Copy.

git switch m2b
cp docs/plans/M2b/files/crates/loopd/tests/device.rs crates/loopd/tests/
cp docs/plans/M2b/files/Makefile Makefile
mkdir -p config && cp docs/plans/M2b/files/config/system.md config/
git diff --stat Makefile

2. The ordinary gate ignores the new tests. make gate. Expected last line: gate: ok, and for device: 0 passed; 0 failed; 6 ignored.
3. The server can be reached. curl -s http://straylight:11434/health. Expected: {"status":"ok"}. If not, stop and report.
4. Run the device checks. make verify-device. Expected: 6 passed and verify-device: ok, in about a minute. The baseline token count is printed; write it in your log row.
5. If a check failed, run it alone with --nocapture (see the M2a task 13 for the command) and put its name and message in your log row with status stopped.
6. Log and commit.

git add Makefile config crates/loopd/tests/device.rs docs/implementer-log.md
git commit

Done when

make gate prints gate: ok and make verify-device prints verify-device: ok.
cmp Makefile docs/plans/M2b/files/Makefile and cmp config/system.md docs/plans/M2b/files/config/system.md print nothing.
git log --oneline master..m2b shows one commit per task.

Stop and report if

The server cannot be reached, or any device check fails twice.

2.2 KiB Raw Blame History