Skip to content

SPEC-078: Review fixtures must stub a review, not a token (re-scope of SPEC-077 after stop-early) #321

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
iselur merged 1 commit into from
Aug 14, 2026
Merged

SPEC-078: Review fixtures must stub a review, not a token (re-scope of SPEC-077 after stop-early) #321

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion tests/review_authorship.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -46,7 +46,7 @@ cat >"$tmp/bin/codex" <<'STUB'
#!/usr/bin/env bash
cat >/dev/null # consume the stdin prompt like the real CLI; never actually invoked in this test
sleep "${CODEX_STUB_SLEEP:-0}"
printf 'stub review verdict\n'
printf '## BLUF\nBottom line: stub review verdict passes the authorship fixture.\nVerified evidence: The local reviewer stub produced this review round.\nOpen findings: None.\nVERDICT: PASS\n'
STUB
chmod +x "$tmp/bin/codex"

Expand Down
2 changes: 1 addition & 1 deletion tests/review_cap.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -34,7 +34,7 @@ cat >"$tmp/bin/codex" <<'STUB'
#!/usr/bin/env bash
cat >/dev/null # consume the stdin prompt like the real CLI
sleep "${CODEX_STUB_SLEEP:-0}"
printf 'stub review verdict\n'
printf '## BLUF\nBottom line: stub review verdict passes the cap fixture.\nVerified evidence: The local reviewer stub produced this review round.\nOpen findings: None.\nVERDICT: PASS\n'
STUB
chmod +x "$tmp/bin/codex"

Expand Down
4 changes: 2 additions & 2 deletions tests/review_kimi.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -30,7 +30,7 @@ chmod u+w "$tmp/repo/scripts/models.json"
cat >"$tmp/bin/codex" <<'STUB'
#!/usr/bin/env bash
cat >/dev/null
printf 'stub codex verdict\n'
printf '## BLUF\nBottom line: stub codex verdict passes the routing fixture.\nVerified evidence: The Codex reviewer path returned this substantive review body.\nOpen findings: None.\nVERDICT: PASS\n'
STUB
chmod +x "$tmp/bin/codex"

Expand All @@ -42,7 +42,7 @@ cat >"$tmp/bin/kimi" <<'STUB'
[[ -z ${KIMI_INVOKED_MARKER:-} ]] || : >"$KIMI_INVOKED_MARKER"
# consume all flags; the real kimi accepts -p/-m/--output-format but we ignore them here
printf '{"role":"user","content":"prompt"}\n'
printf '{"role":"assistant","content":"stub kimi verdict"}\n'
printf '{"role":"assistant","content":"## BLUF\\nBottom line: stub kimi verdict passes the routing fixture.\\nVerified evidence: The Kimi reviewer path returned this substantive review body.\\nOpen findings: None.\\nVERDICT: PASS"}\n'
STUB
chmod +x "$tmp/bin/kimi"

Expand Down
2 changes: 1 addition & 1 deletion tests/review_plan_authorship.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,7 +38,7 @@ cat >"$tmp/bin/codex" <<'STUB'
#!/usr/bin/env bash
cat >/dev/null # consume the stdin prompt like the real CLI
sleep "${CODEX_STUB_SLEEP:-0}"
printf 'stub review verdict\n'
printf '## BLUF\nBottom line: stub review verdict passes the plan-authorship fixture.\nVerified evidence: The local reviewer stub produced this review round.\nOpen findings: None.\nVERDICT: PASS\n'
STUB
chmod +x "$tmp/bin/codex"

Expand Down
4 changes: 2 additions & 2 deletions tests/vendor_cli_parity.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -397,8 +397,8 @@ run_review() {
}

rm -rf "$tmp/review-base/.orchestrator" "$tmp/review-cli/.orchestrator"
printf 'codex review answer\n' >"$tmp/codex-review-output"
printf '{"role":"assistant","content":"kimi review answer"}\n' >"$tmp/kimi-review-output"
printf '## BLUF\nBottom line: The Codex review fixture passes.\nVerified evidence: The parity path returned this substantive review body.\nOpen findings: None.\nVERDICT: PASS\n' >"$tmp/codex-review-output"
printf '{"role":"assistant","content":"## BLUF\\nBottom line: The Kimi review fixture passes.\\nVerified evidence: The parity path returned this substantive review body.\\nOpen findings: None.\\nVERDICT: PASS"}\n' >"$tmp/kimi-review-output"

# With high effort, Codex argv, stdin, recovered answer, and observable status stay byte-identical.
configure_review codex high
Expand Down
Loading