fix(opencode): keep oversized reviews packet-first

[seonghobae]

Summary

• repeat key current-head evidence sections in the inline OpenCode packet so oversized PRs still expose changed files, focused hunks, coverage, failed checks, and unresolved threads when direct file reads do not execute
• tighten the OpenCode prompt and launcher to forbid raw tool-call markup / function-call JSON in review bodies
• clarify that tool/file access gaps are source limitations, not PR blockers, unless current-head evidence reports a materialization failure; REQUEST_CHANGES findings must cite positive lines, never line 0

Evidence

• git diff --check
• bash -n scripts/ci/run_opencode_review_model_pool.sh
• bash scripts/ci/strix_required_workflow_smoke.sh
• uv run python -m py_compile scripts/ci/pr_review_merge_scheduler.py scripts/ci/sandboxed_web_e2e.py scripts/ci/review_execution_contracts.py
• uv run pytest tests/test_opencode_agent_contract.py tests/test_review_execution_contracts.py tests/test_required_workflow_queue_contract.py tests/test_pr_review_merge_scheduler.py tests/test_sandboxed_web_e2e.py

[opencode-agent]

OpenCode Review Overview

• Head SHA: ec5394852c3377246d2224e61af923671d96740d
• Workflow run: 29143886120
• Workflow attempt: 1
• Gate result: APPROVE (exit 0)

Changed-File Evidence Map

flowchart LR
  PR["PR changed files"] --> Evidence["OpenCode bounded evidence"]
  Evidence --> S1["Workflow: opencode-review.yml"]
  S1 --> I1["GitHub Actions review job"]
  I1 --> R1["Review risk: Workflow: opencode-review.yml"]
  R1 --> V1["actionlint plus required checks"]
  Evidence --> S2["CI script (2 files)"]
  S2 --> I2["review and security gate shell path"]
  I2 --> R2["Review risk: CI script (2 files)"]
  R2 --> V2["bash -n plus Strix self-test"]
  Evidence --> S3["Test: test_opencode_agent_contract.py"]
  S3 --> I3["regression suite"]
  I3 --> R3["Review risk: Test: test_opencode_agent_contract.py"]
  R3 --> V3["targeted test run"]

Loading