SuperAPI AI Directory

gs-council

Coding & development

View repository

Updated 11-19 18:05

About this skill

Council

Before an audit, address outstanding PR comments, fix known issues, and run relevant checks. Give council the updated implementation; do not defer known fixes until reviews return. If the user wants to discard the implementation and explore new ideas, skip fixes to code being replaced.

Bring in three delegates to help with the user's request, usually for review or feedback. Use one strong model from each of three different providers, at high reasoning effort when the harness lets you set it. If the user's instructions name council members, use those. Otherwise pick the latest flagship model from each provider the harness reaches, for example Claude Opus, GPT Sol, and Gemini Pro.

Use the harness's subagent tools to start one delegate per model in parallel. In T3 Code, for example, use orchestrator_capabilities to resolve the models, then delegate_task. If a named member is unavailable, report it instead of substituting. If the harness reaches fewer than three providers, say which are missing, then fill the remaining seats with the strongest models it does reach. If the harness has no subagent tools, tell the user and stop.

Give each delegate the request and relevant context. For review or feedback, keep their work read-only. Collect their responses and summarize the useful findings, disagreements, and recommended next steps.

Do not wait idle. When a result arrives, check its findings against the code at once and decide what to fix while the other delegates run. Do not edit files while a delegate still reviews the same checkout. If the user's instructions mark a delegate as non-blocking, do not wait for it; treat a late result as feedback and do not record its score.

For a code review, ask each delegate for a score out of 10. Count each full review pass as one round. A fresh review after fixes starts a new round.

If the work ends in a pull request, put the results in the PR description. If the description ends with a line naming the model and harness, put this section before it:

## Council review

Rounds: 2

| Delegate | Round 1 | Round 2 |
| --- | --- | --- |
| Claude Opus | 7/10 | 9/10 |
| GPT Sol | 8/10 | 10/10 |
| Gemini Pro | 7/10 | 9/10 |

If a delegate gives no score, write no score. Do not invent one. If the PR is open, edit the description. Do not post a comment for this. If there is no PR, put the same results in the final reply only.

If you agree with their recommendations, make the changes and request a fresh review after. Continue this process until all delegates are satisfied with the outcome.

Do not let repeated reviews result in scope creep. It is your job to fight to keep scope limited.