Problem
The Gemini consult lane (Antigravity agy, migrated in #778) runs successfully but does not produce a structured review verdict when asked for one (consult -m gemini --type <spec|plan|impl|phase|pr|integration>). Instead of returning VERDICT: APPROVE/REQUEST_CHANGES with findings, it wanders off-task — exploring the repo and, notably, latching onto the local google-antigravity-sdk skill and summarizing that instead of reviewing the target.
Because no verdict comes back, porch's review loop treats it as REQUEST_CHANGES (see companion porch bug) and loops every phase. This is not an auth problem — agy 1.0.6 authenticates fine (OAuth, exit 0); the lane is behaviorally broken for structured reviews.
Reproductions
- Architect smoke test (2026-06-09):
consult -m gemini --prompt "Reply with exactly: AGY_LANE_OK <year>. Nothing else." → completed in 47.6s, exit 0, but ignored the instruction and produced a multi-section "sandbox state" summary that referenced google-antigravity-sdk SKILL.md. Agentic exploration overrode the prompt.
- Builder spir-987, phase_1 (2026-06-11/12): iter1 timed out (skipped, fine); iter2 ran but went off-task — explored an unrelated
google-antigravity-sdk, returned no VERDICT. porch defaulted to REQUEST_CHANGES → loop (reached iter3 of max_iterations=8). Codex + Claude both APPROVE'd the same phase cleanly (build + 5 tests green).
Suspected root cause
agy's agentic/--sandbox mode lets Gemini self-direct, and the presence of the google-antigravity-sdk skill in context appears to hijack it (it keeps exploring that). The review prompt's "return a VERDICT" instruction is not being honored / not constraining the run.
Expected
- A
--type review consult on the gemini lane returns a parseable VERDICT (APPROVE / REQUEST_CHANGES + findings), same contract as codex/claude.
- agy should be constrained to the review task (less free-roaming agentic exploration for
--type reviews), and should not be derailed by unrelated local skills.
- If the lane genuinely can't produce structured verdicts reliably, consider gating gemini out of the default review model set until it can.
Impact / workaround
Surfaced on #987; worked around by scoping that project's consult models to codex + claude only. Left unfixed, this recurs on every phase of every project that uses the default 3-model review set. Related: #778 (the agy migration), and the companion porch no-verdict bug (filed alongside this).
Problem
The Gemini consult lane (Antigravity
agy, migrated in #778) runs successfully but does not produce a structured review verdict when asked for one (consult -m gemini --type <spec|plan|impl|phase|pr|integration>). Instead of returningVERDICT: APPROVE/REQUEST_CHANGESwith findings, it wanders off-task — exploring the repo and, notably, latching onto the localgoogle-antigravity-sdkskill and summarizing that instead of reviewing the target.Because no verdict comes back, porch's review loop treats it as
REQUEST_CHANGES(see companion porch bug) and loops every phase. This is not an auth problem —agy1.0.6 authenticates fine (OAuth, exit 0); the lane is behaviorally broken for structured reviews.Reproductions
consult -m gemini --prompt "Reply with exactly: AGY_LANE_OK <year>. Nothing else."→ completed in 47.6s, exit 0, but ignored the instruction and produced a multi-section "sandbox state" summary that referencedgoogle-antigravity-sdkSKILL.md. Agentic exploration overrode the prompt.google-antigravity-sdk, returned no VERDICT. porch defaulted toREQUEST_CHANGES→ loop (reached iter3 ofmax_iterations=8). Codex + Claude bothAPPROVE'd the same phase cleanly (build + 5 tests green).Suspected root cause
agy's agentic/--sandboxmode lets Gemini self-direct, and the presence of thegoogle-antigravity-sdkskill in context appears to hijack it (it keeps exploring that). The review prompt's "return a VERDICT" instruction is not being honored / not constraining the run.Expected
--typereview consult on the gemini lane returns a parseableVERDICT(APPROVE / REQUEST_CHANGES + findings), same contract as codex/claude.--typereviews), and should not be derailed by unrelated local skills.Impact / workaround
Surfaced on #987; worked around by scoping that project's consult models to codex + claude only. Left unfixed, this recurs on every phase of every project that uses the default 3-model review set. Related: #778 (the agy migration), and the companion porch no-verdict bug (filed alongside this).