skill-selection

skill-selection is a skill for Claude Code from anhnguyen0905/codex-mcp. It costs 87 tokens per session (5,448 once invoked), scanned B, original, MIT.

A skill-selection guide for choosing only the task-specific instructions from a local catalogue of available skills.

In plain words
What is it for?
It helps classify a request, match it to local or remote skills, check whether the index is current, and load the skills that fit the available context.
Why use it?
It prevents loading unrelated instructions that consume context and make the relevant guidance harder to find.

Skill for Claude Code

Written for Claude Code: ${CLAUDE_PLUGIN_ROOT} variable. Also seen: reads .claude/ paths; names the AskUserQuestion tool; mentions Codex.

Runs only inside its plugin — its command needs a path that Claude Code sets for a plugin’s own hooks and for nothing else. Install the plugin, not this.

Part of the codex-flow plugin — 61 skills, 1 command, 1 MCP server shipped together

Good fit It helps classify a request, match it to local or remote skills, check whether the index is current, and load the skills that fit the available context.

Compare 6 skills from other repositories ↓
Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code
/plugin marketplace add anhnguyen0905/codex-mcp
Claude Code
/plugin install codex-flow

Made for: Claude Code.

Or install codex-flow, the plugin that ships this one along with the rest of its 61 skills, 1 command, 1 MCP server.

Wrote 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.

agentmods badge for skill-selection

README.md
[![agentmods](https://agentmods.dev/badge/skills/anhnguyen0905/codex-mcp/skill-selection.svg)](https://agentmods.dev/skills/anhnguyen0905/codex-mcp/skill-selection)
Your own site
<a href="https://agentmods.dev/skills/anhnguyen0905/codex-mcp/skill-selection"><img src="https://agentmods.dev/badge/skills/anhnguyen0905/codex-mcp/skill-selection.svg" alt="Measured on agentmods" height="20"></a>
Per session 87 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 5,448 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00087 $0.05448
Opus 5 $0.00044 $0.02724
Sonnet 5 $0.00017 $0.01090
Haiku 4.5 $0.00009 $0.00545

Measured 7d ago against content hash 8b05bb09adf7, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade B, and why

skill-selection scanned grade B with 1 finding 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.

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

`grep -ril "<term>" ~/.claude/plugins/cache/*/*/*/skills --include=SKILL.md`
skills/skill-selection/SKILL.md · 311 lines

How it starts

The opening of the file, as written. The whole thing — 311 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Skill Selection (index-first, never blind-load)

Loading every available skill wastes context and buries the ones that matter. Select the few relevant skills from a local index; treat "0 matches" as a normal outcome. A session should only ever contain the skills its current work needs.

Step 1 — Locate the index

  • Path: $CODEX_FLOW_SKILLS_INDEX if set, else ~/.claude/skill-library/INDEX.md.
  • Missing or stale (skills were added since it was rebuilt)? Rebuild it: node "${CLAUDE_PLUGIN_ROOT}/scripts/build-skills-index.mjs" [extra skill dirs ...] — defaults scan ~/.claude/skills, ~/claude-skill-library (including promoted skills under remote/), and the skills/ dir of every installed plugin (~/.claude/plugins/cache/<marketplace>/<plugin>/<newest version>/skills), then merge the REMOTE.md pointer catalog for anything not yet local. Installed plugin skills are trusted and load by name — an index that omits them makes selection claim a gap the machine already fills.
  • Sanity-check the count before trusting a "no match": an index with only a few dozen entries almost certainly predates the plugin scan. Rebuild, then match again. Anything under a quarantine/ directory (where --clone lands third-party repos) is NEVER indexed — quarantined skills only become visible through the explicit vet step in Step 5.
  • Still no index → skip selection, continue the phase with its named codex-flow:* skills only, and tell the user once how to enable the local-first setup: node "${CLAUDE_PLUGIN_ROOT}/scripts/sync-awesome-skills.mjs" --clone then rebuild the index.
  • The library dir (~/claude-skill-library) must stay OUTSIDE auto-discovered skill dirs — skills on disk cost zero context until selected; only what you load enters the session.

Step 2 — Classify the request, then derive search terms

Requests are not always engineering. First identify 1–2 role facets the task actually needs:

Facet Examples of terms it generates
Engineering language (python, typescript…), framework (django, react…), storage (postgres, redis…), domain (auth, payments, migrations, e2e, performance…)
Data / analytics pandas, statistics, cohort, retention, funnel, segmentation, forecasting, experiment, significance
Data engineering sql, warehouse, star schema, dbt, etl, orchestration, event taxonomy, data quality, reconciliation
Visualization / reporting chart, palette, dashboard, slides/pptx, spreadsheet/xlsx, infographic, executive summary
Marketing / content campaign, positioning, messaging, copywriting, seo, landing page, social, brand voice, content calendar, launch
Growth / paid media roas, cpi, cac, ltv, payback, attribution, incrementality, bidding, media plan, creative fatigue, aso
Research market sizing, competitor, benchmark, survey design, cross-tab, persona, interview, scenario / what-if
Product / planning prd, roadmap, backlog, prioritization, metrics, kpi tree, okr, pricing, unit economics
Finance / BizOps business case, budget allocation, forecast, margin, sensitivity, sop
Accounting / audit bookkeeping, journal entry, reconciliation, chart of accounts, accrual, month-end close, P&L, balance sheet, VAT, invoice, trial balance, audit, materiality, sampling, internal controls, workpapers
Personal finance household budget, emergency fund, debt payoff, savings rate, compound interest, retirement, net worth, insurance
SME / operations small business, pricing, break-even, cash flow, runway, working capital, supplier terms, inventory, sop, production scheduling, capacity, quality
Sales / support / CRM pipeline, lead qualification, quota, forecast, win/loss, crm, ticket triage, sla, csat, nps, escalation, knowledge base
HR / people recruiting, job description, structured interview, onboarding, compensation, performance review, engagement survey, attrition, headcount
Legal / compliance contract review, nda, clause, termination, liability, indemnification, redline, compliance, data privacy, customs
Design / UX ui, ux research, persona, usability testing, user flow, wireframe, information architecture, accessibility, motion, design system, typography
Education / training learning objectives, curriculum, syllabus, lesson plan, assessment, rubric, e-learning, workshop, skill gap
Real estate rental yield, cap rate, noi, cash-on-cash, mortgage, amortization, buy vs rent, occupancy, vacancy
Logistics / supply chain carrier, freight, customs, hs code, incoterms, shipment exception, returns, reverse logistics, demand planning, safety stock, warehouse
Localization translation, locale, tone preservation, market adaptation

Read the full file on GitHub · 311 lines

Changes

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.

  1. 7d ago First seen · 311 lines · 87 tokens per session scan B 8b05bb09adf7

Subscribe to this mod's changes

skill-selection is a skill published in the GitHub repository anhnguyen0905/codex-mcp (3 stars, last pushed 5d ago), licensed MIT. It adds 87 tokens to every session and 5,448 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

local-ai-agents

Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…

microsoft/ai-agents-for-beginners · 200 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

next-partial-prefetching-adoption

Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…

vercel/next.js · 103 tokens

chronicle

Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…

microsoft/vscode · 72 tokens

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens