Advisor review

Advisor Review

Use the bundled scripts from this skill directory. An explicit invocation is a latch: run the reviewer in the same parent task before a substantive final answer or an action that follows or conflicts with the review. Read-only evidence collection may continue while constructing the packet.

Do not search the global tool catalog for an advisor. Do not open a second user-visible task. The only child execution is the isolated runtime CLI process started by scripts/run_advisor.py.

Requirements

Require shell execution, an authenticated CLI for the active session runtime, access to the caller-selected advisor model, and outbound access for that runtime from the parent shell. Supported runtimes are codex, claude, opencode, and gemini; --runtime auto uses the active session marker such as OPENCODE=1, and refuses an ambiguous host instead of guessing. A parent sandbox that blocks subprocess networking also blocks the nested reviewer. If any requirement is missing, report that the independent review is blocked; do not weaken the sandbox without user or task authority.

The runner is not a confidentiality boundary. Sanitize evidence and artifacts before building the packet.

Workflow

1. Announce that Advisor Review is being used and why.

2. Select one phase:

   • plan: challenge a proposed approach.
   • stuck: diagnose failed or looping work.
   • pivot: compare a materially different approach.
   • final: audit completed work and its endpoint evidence.

3. Read references/context-contract.md.

4. Choose the context mode:

   • packet: short source-anchored facts are sufficient.
   • bundle: sanitized logs, timelines, diffs, or other bounded text artifacts materially affect the diagnosis.

5. Build advisor-context-2.0 with scripts/build_context_packet.py.

   • For stuck, include actions and exact results in chronological order.
   • Give every observed fact a concrete source.
   • Include negative results, conflicting state, and context limitations.
   • Never silently truncate decision-critical text. A builder failure is a blocked review.

6. Read references/review-rubric.md.

7. Select the advisor model and effort as the calling agent. Use a model available to the selected runtime, honor an explicit model or effort request, and otherwise choose based on the review phase and risk:

   1. Use max only for an explicitly requested deepest review or a critical, hard-to-reverse decision with major unresolved ambiguity.
   2. Use xhigh for cross-component failures, conflicting verified evidence, two or more failed approaches, major pivots, or high-impact completion claims.
   3. Otherwise use high.

8. Run exactly one isolated review by default and save its receipt. Pass the active session runtime, selected model, and selected effort explicitly so the receipt records the caller's request:

    python3 scripts/run_advisor.py \
      --input context-packet.json \
      --runtime auto \
      --model <caller-selected-model> \
      --effort xhigh \
      --output advisor-receipt.json
   

   Keep the returned receipt in the invoking task. A receipt produced in another user-visible task is not a handoff.

9. Check the receipt:

   • request.runtime must be the active session runtime or the explicitly selected runtime.
   • request.requested_model must equal the model selected by the calling agent.
   • request.requested_effort must match the selected effort.
   • observed_model and observed_effort may be null; runtime output does not authoritatively expose serving-side identity. Never describe requested identity as verified actual identity.
   • Any runner error, schema failure, timeout, or context-hash mismatch blocks the review.
   • The isolated child uses read-only or plan-mode controls, disables session persistence where supported, and disables plugin/skill loading where the runtime exposes that control. The Codex adapter uses a temporary CODEX_HOME with only the host authentication file linked when present; OpenCode runs with --pure and external skill/config loading disabled.

10. Read references/decision-contract.md. Resolve every R# recommendation as adopt, reject, or defer, then run:

    python3 scripts/validate_decision.py \
      --receipt advisor-receipt.json \
      --decision advisor-decision.json
    

11. Do not act on advisor recommendations until the decision record validates. A destructive recommendation additionally requires confirmed user authority.

12. Continue the parent task using the validated next action and stop condition.

Stuck-work rules

• Diagnose before changing more state.
• Distinguish the execution boundary: parent task, isolated reviewer process, and any other user-visible task are separate.
• Do not let a separate task's receipt stand in for a same-task handoff.
• Prefer one read-only or reversible experiment that separates competing hypotheses.
• After a repeated user correction, do not repeat the same prescription without new evidence.
• Do not broaden scope from one failing subsystem to plugins, services, repositories, or packages unless evidence connects them.

Review budget

Use one review by default. A second is allowed only when new evidence materially changes the decision or a high-risk final re-check is justified. A third is allowed only to reconcile a concrete conflict. Never create an open-ended review loop.

Skill-change behavior gate

When modifying Advisor Review itself, run the deterministic tests and at least one sanitized behavior fixture:

python3 scripts/evaluate_advisor_behavior.py \
  --fixture behavior-fixture.json \
  --receipt advisor-receipt.json

Use paired old/new fixtures where practical. Judge detection, evidence linkage, bounded experiments, and risky-action rate—not prose similarity.

Reporting

Report:

• phase, context mode, selected runtime, requested advisor model, requested effort, and identity verification status;
• advisor verdict and evidence-linked diagnosis;
• each consequential recommendation's adopt/reject/defer disposition;
• the selected next action and stop condition;
• unresolved risks or context limitations.

Call the result an advisor-style independent review, not parity with any one runtime's native advisor feature.