Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add xavierchoi/claude-plugin-codex --skill consult-claudegit clone --depth 1 https://github.com/xavierchoi/claude-plugin-codexWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/xavierchoi/claude-plugin-codex/consult-claude)<a href="https://agentmods.dev/skills/xavierchoi/claude-plugin-codex/consult-claude"><img src="https://agentmods.dev/badge/skills/xavierchoi/claude-plugin-codex/consult-claude.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00089 | $0.01758 |
| Opus 5 | $0.00044 | $0.00879 |
| Sonnet 5 | $0.00018 | $0.00352 |
| Haiku 4.5 | $0.00009 | $0.00176 |
Grade A, and why
consult-claude scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 7d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 138 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Consult Claude Code
This plugin lets you hand work to Claude Code as a collaborative second agent. Claude runs locally in the same repository (reusing the user's existing Claude login) and reports back.
Keep the spirit gentle and collaborative: Claude is a peer offering a careful second perspective, not a last resort or an adversary.
First-time setup
If consulting fails, or the user asks whether the plugin is ready, call the
setup tool first. It checks that claude is installed and signed in and
returns any next steps (it reuses the user's existing Claude login — no extra
API key needed). Pass deep: true to verify the login with a quick live call.
If the tools themselves fail (transport closed)
If claude-code tool calls fail with a transport/connection error (e.g.
"Transport closed"), the MCP bridge for this session is gone — this happens
when the plugin was upgraded while the session was open. Tell the user to
start a fresh Codex session; the tools will work again there. Do not
work around it by invoking the claude CLI directly — especially not with
permission overrides like --dangerously-skip-permissions; the bridge exists
precisely to run Claude with safe, scoped permissions.
How to call it
Use the consult tool from the claude-code MCP server. Arguments:
prompt— what you'd like Claude to look at, investigate, or do. Be specific about the goal and what "done" looks like.cwd— always pass the absolute path of the repository you are working in. The tool cannot infer it.edit—truelets Claude modify files;false(default) is advisory only.background—truereturns a job id immediately and runs in the background; poll withconsult_status, fetch withconsult_result, stop withconsult_cancel.verify— a command the server runs after the edits (e.g."auto"or"npm test"); see below.resume— continue the most recent Claude session for this directory.model/effort— optional model (alias likesonnet/opusor a full name) and effort level (low…max); see the inference rules below.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 7d ago First seen · 138 lines · 89 tokens per session scan A 27db19203bfc
consult-claude is a skill published in the GitHub repository xavierchoi/claude-plugin-codex (0 stars, last pushed 2mo ago), licensed MIT. It adds 89 tokens to every session and 1,758 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
watch
File sentinel that monitors the working directory for changes and marker comments, then auto-triggers appropriate skills. Poll-based via git diff against the last scan commit. Writes intake items for batch processing and routes marker actions through /do. Use for automatic reactions to file changes; do NOT use for…
review
5-pass structured code review — correctness, security, performance, readability, consistency.
live-preview
Mid-build visual verification loop. Takes screenshots of components during construction, not just after. Catches visual regressions and invisible features before they compound. Requires Playwright or similar screenshot tool.
marshal
Meta-orchestrator that takes any direction — broad, specific, or vague — and autonomously chains skills and context into actionable work. Gathers context from codebase, docs, and memory. Only asks the user when it genuinely cannot proceed. Single-session orchestrator.
wiki
Markdown-first knowledge base where the LLM acts as librarian. Ingests raw sources, compiles and interlinks topic files, self-maintains an index. No vector DB or embeddings required -- uses LLM-native navigation over structured markdown up to 400K words.
stss
Reduce defensive disclaimers, stacked hedging, and self-protective narration in proposals and decision-facing writing. Use when the user asks to rewrite or audit a proposal, plan, research contribution, executive summary, or similar text for directness. Do not use for ordinary code work or unrelated prose.