refactor the skills to reduce tokens and improve progresive disclousure patterns

[Cesar-M-Diaz]

No description provided.

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on base/target branches other than the default branch.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro

Run ID: 9a339f4d-cb7d-4a4d-b7ca-410df2d38bd0

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

---

_{Comment @coderabbitai help to get the list of available commands.}

[dobleuber]

I think we have a conflict in the skills instructions

[dobleuber]

The new wait step instructs agents to run node "<skill-dir>/wait.cjs" for CPU/heap asset generation, but this same skill still says to “Use ONLY N|Solid MCP tools” and “Do not use shell commands or bundled helper scripts for waits or downloads” in the guardrails. In practice, an agent following the stricter guardrail can refuse the required wait step, while an agent following step 4 violates the guardrail; this makes CPU/heap asset capture workflows unreliable. Please update the guardrails to allow the bundled wait helper, or switch the wait step back to the intended MCP wait tool.

[jdaison]

LGTM