agentsclimarketplace

Session baseline

Skill adeilsonrbrito/cross-model-consult/session-baseline

Cross-model second opinions for coding agents — skills that let Claude Code consult rival-model CLIs (Codex, Grok, OpenCode), solo or as a blind three-model panel.

Install
npx -y skills add adeilsonrbrito/cross-model-consult --skill session-baseline

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • 25 days oldThe repository was created 25 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Baseline operating posture for a Claude Code session — subagent delegation and self-directed cross-model consults. Opt-in and task-agnostic: invoke it at the start of a session to set these defaults, then layer a task-specific skill (discussion, planning, implementation) on top. Use whenever you want the standard operating rules regardless of task complexity.

SKILL.md

1.4 KB, 200 tokens by cl100k_base, as published. Nobody here has run it

Session baseline

These are session-wide operating rules. Once set, they remain in force for the entire session — across whatever task-specific skills (discussion, planning, implementation) you layer on top.

Delegate to subagents. Whenever possible, use subagents for tasks like investigation, exploration, etc., aiming to not bloat the context window. When spawning subagents, prefer the strongest model available in the environment at high reasoning effort.

Consult your cross-model team. You may self-invoke consult-codex, consult-opencode, and consult-grok without asking the user first — treat them as your team of partners. Follow each consult skill's own cost bar: reach for them on genuinely hard, high-stakes, or ambiguous calls where cross-model divergence would change the outcome, not as a default reflex for routine decisions. Prefer consult-panel only when a single outside opinion is not enough. Defer to any per-project policy that restricts sending context to external providers.

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.