consult-claude

Give Claude one fresh, read-only adversarial consultation on Codex's current synthesis. Use only when the user explicitly invokes $consult-claude or unmistakably asks Codex to consult Claude on an uncompromising greenfield vision, architecture direction, or other bounded decision after Codex has gathered the relevant evidence and directional data.

Quality

95%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Quality

Content

100%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The content is an efficient, fully actionable runbook: concrete packet templates, exact execution mechanics, a clearly sequenced workflow with verification and cancellation checkpoints, and a clean single-file structure with one real script reference. No dimensions warrant deduction.

Dimension	Reasoning	Score
Conciseness	The body is lean and purposeful — it provides the exact packet templates and run mechanics without explaining generic concepts Claude already knows, and every section earns its tokens.	3 / 3
Actionability	Gives copy-paste-ready packet templates (Mandate, Mission/Evolution/Evidence/etc.) and concrete executable run guidance (Bun, scripts/consult-claude.ts, PTY, raw non-echoing stdin, EOT/Ctrl-D), with a real referenced script.	3 / 3
Workflow Clarity	A clear four-phase sequence (build packet -> run consultation -> wait patiently -> reconcile memo) with explicit checkpoints, including a verify-claims feedback loop in reconciliation and timeout/cancel handling in the wait phase.	3 / 3
Progressive Disclosure	A single well-organized SKILL.md split into clearly signaled sections, with one one-level-deep bundle reference (scripts/consult-claude.ts) that exists on disk and no nested reference chains.	3 / 3
	Total	12 / 12 Passed

Description

90%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A tight, third-person description that clearly states the single action and gives an explicit, natural trigger clause tied to a distinctive niche. Its only weakness is specificity: it names one action rather than enumerating the concrete steps of the consultation.

Suggestions

Add the concrete deliverable to the action, e.g. '...consultation on Codex's current synthesis and returns one decisive adversarial memo', so the description lists the input-to-output actions rather than a single action.

Dimension	Reasoning	Score
Specificity	Names the concrete domain and action ('one fresh, read-only adversarial consultation on Codex's current synthesis') but describes a single action rather than listing multiple specific concrete actions, so it stops short of the top anchor.	2 / 3
Completeness	States both what it does ('Give Claude one fresh, read-only adversarial consultation') and when to use it with an explicit 'Use only when the user explicitly invokes...' trigger clause, satisfying the top anchor.	3 / 3
Trigger Term Quality	Covers natural trigger phrasings a user in this workflow would say — 'explicitly invokes $consult-claude', 'asks Codex to consult Claude', 'greenfield vision', 'architecture direction', and 'bounded decision'.	3 / 3
Distinctiveness Conflict Risk	The $consult-claude invocation and Codex-specific consultation framing occupy a narrow, distinctive niche with explicit triggers unlikely to fire for unrelated skills.	3 / 3
	Total	11 / 12 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: EpicenterHQ/epicenter
Path: .agents/skills/consult-claude/SKILL.md
Commit: cb12bcc

Reviewed: 3 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.