skills CLAUDE.md

skills CLAUDE.md is an instructions file for coding agents from pinecone-io/skills. It costs 517 tokens per session, scanned A, original, MIT.

Repository instructions for creating reusable coding-agent skills in the pinecone-io/skills project. They explain how skills are authored, adapted for different agents, and checked before release.

In plain words
What is it for?
Use them when editing or reviewing skills, target-specific instructions, manifests, or related documentation in that repository.
Why use it?
They prevent repository-specific conventions and agent compatibility requirements from being missed.

Instructions file

Install

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.

agentmods
npx agentmods add instructions/pinecone-io/skills/claude-md
Clone the repo
git clone --depth 1 https://github.com/pinecone-io/skills

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 skills CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/pinecone-io/skills/claude-md.svg)](https://agentmods.dev/instructions/pinecone-io/skills/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/pinecone-io/skills/claude-md"><img src="https://agentmods.dev/badge/instructions/pinecone-io/skills/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 517 This file is loaded in full into every session.
When invoked 517 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
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 $0.00517 $0.00517
Opus 5 $0.00259 $0.00259
Sonnet 5 $0.00103 $0.00103
Haiku 4.5 $0.00052 $0.00052

Measured 5d ago against content hash 72d0d3b7ca35, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

skills CLAUDE.md 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 5d 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.

CLAUDE.md · 34 lines

How it starts

The opening of the file, as written. The whole thing — 34 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Skills Repo — Authoring Guidelines

This is the base skills repo and the only place skills are authored. Skills here must work across any agent environment (Claude Code, Cursor, etc.).

Plugin repos receive generated output. tools/build.py renders this tree per target using the manifests in targets/, and the sync workflow opens a pull request against each plugin repo. Nothing rewrites skills after they leave here, so what you write is what ships. Where a target genuinely needs different wording, base leaves a <<marker>> and each file in targets/ supplies its own text — see targets/README.md.

Generalization Rules

No IDE-specific tool calls in base. Don't name AskUserQuestion or any other tool that isn't universally available. Write plain prose: "ask the user which they prefer."

Where a tool really is the right answer for one target, use a snippet instead of dropping the idea. Base leaves a <<marker>>; each file in targets/ supplies its own wording. That is how AskUserQuestion reaches the Claude Code plugin while Cursor gets prose — see targets/README.md → Snippets. Earlier the rule was a flat ban, which quietly cost the Claude plugin a working feature at 15 sites.

Add a snippet only where the targets truly differ. If one wording is right everywhere, it belongs in base.

Never assume PINECONE_API_KEY is inherited. Always check first. If not set:

  • Tell the user to export PINECONE_API_KEY=... (terminal envs)
  • Or create a .env file and run scripts with uv run --env-file .env scripts/... (IDE envs that don't inherit shell vars)

MCP-dependent skills must say so. If a skill requires MCP, state it in the description and fail fast with a helpful message if tools aren't available.

No IDE-specific language. Avoid phrases like "Claude Code console" or references to specific IDE UI.

No fake API key literals. Never write pcsk_... or similar patterns — use your-key as a placeholder. Security linters will flag fake key patterns.

Read the full file on GitHub · 34 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. 5d ago First seen · 34 lines · 517 tokens per session scan A 72d0d3b7ca35

Subscribe to this mod's changes

skills CLAUDE.md is an instructions file published in the GitHub repository pinecone-io/skills (15 stars, last pushed yesterday), licensed MIT. It adds 517 tokens to every session, about $0.0026 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 2026-08-30.