Record the GPT-5.6 Luna SCV receipt

Record the GPT-5.6 Luna SCV receipt

Deploy story

What this commit did to the running system — joined from the forge receipt chain, the part a commit page elsewhere cannot show.

Not deployed through the forge lane

No push, promotion, build, or deploy receipt references this commit (receipts are scanned over a bounded recent window). Changes shipped by full node replacement carry their proof in the release gate receipt instead.

Changed files

modified docs/operations/scv-staging-qualification.md
modified docs/scv-planning.md

Diff

2 files changed, +35 -0

101	101		`two read calls, eight normalized events, 754,118,656 bytes of peak RSS, 45`
102	102		`resource samples, no sampling errors, and no output truncation. This second`
103	103		`receipt proves the currently deployed job configuration.`
	104	+
	105	+	`## GPT-5.6 Luna staging receipt`
	106	+
	107	+	Cloud Build `655a2fe6-8173-4686-9533-f3a1733942b0` built source revision
	108	+	`c7175ed8a8781ea1aab7d204623a28ac45a70bc1`. Artifact Registry resolved the
	109	+	`worker image to`
	110	+	`sha256:156ff9f51e955d03b4795af8e2bb190a6c4f9962cc7942a6dcc0586e9b48b0a9`.
	111	+	`The control-plane, parity, production-readiness, and staging SCV jobs all use`
	112	+	that digest with `SCV_MODEL=openai/gpt-5.6-luna` and
	113	+	`SCV_REASONING_EFFORT=low`.
	114	+
	115	+	Parity execution `openagents-scv-parity-audit-p5mkz` completed successfully as
	116	+	SCV run `0f13d278-02aa-47cd-bc07-3bdcd2fd6135`. The OpenCode process ran for
	117	+	58,553 milliseconds and emitted 62 normalized events. It completed 9 `glob`,
	118	+	9 `grep`, and 22 `read` calls, recorded 688,361,472 bytes of peak RSS, and
	119	+	`reported no sampling or output truncation.`
	120	+
	121	+	`Cloud Logging stored the terminal result as`
	122	+	`openagents.scv.worker.result.v1` and the 8,645-byte report as three ordered
	123	+	`openagents.scv.report.chunk.v1` JSON records. Both the initial
	124	+	`run_preparing` event and terminal result recorded the exact model and reasoning
	125	+	`effort. This receipt qualifies the Luna model-selection, live event, bounded`
	126	+	`report, and terminal-result paths in the shared staging project. It does not`
	127	+	`admit repository writes or count the report findings as claimed or completed`
	128	+	`issues.`

this plan and does not authorize writes, Forge handoff, or autonomous
deployment.

The current staging worker now uses `openai/gpt-5.6-luna` with `low` reasoning.
Cloud Build `655a2fe6-8173-4686-9533-f3a1733942b0` built revision
`c7175ed8a8781ea1aab7d204623a28ac45a70bc1` as image digest
`sha256:156ff9f51e955d03b4795af8e2bb190a6c4f9962cc7942a6dcc0586e9b48b0a9`.
Parity execution `openagents-scv-parity-audit-p5mkz` proved the exact model and
reasoning effort in its live preparation event and terminal result. Its bounded
report arrived as three ordered structured JSON chunks without truncation. See
[Qualify an SCV in staging](operations/scv-staging-qualification.md) for the
complete receipt.

## Goals

An SCV should:

300	300		`this plan and does not authorize writes, Forge handoff, or autonomous`
301	301		`deployment.`
302	302
	303	+	The current staging worker now uses `openai/gpt-5.6-luna` with `low` reasoning.
	304	+	Cloud Build `655a2fe6-8173-4686-9533-f3a1733942b0` built revision
	305	+	`c7175ed8a8781ea1aab7d204623a28ac45a70bc1` as image digest
	306	+	`sha256:156ff9f51e955d03b4795af8e2bb190a6c4f9962cc7942a6dcc0586e9b48b0a9`.
	307	+	Parity execution `openagents-scv-parity-audit-p5mkz` proved the exact model and
	308	+	`reasoning effort in its live preparation event and terminal result. Its bounded`
	309	+	`report arrived as three ordered structured JSON chunks without truncation. See`
	310	+	`[Qualify an SCV in staging](operations/scv-staging-qualification.md) for the`
	311	+	`complete receipt.`
	312	+
303	313		`## Goals`
304	314
305	315		`An SCV should:`