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 agentmods add instructions/pinecone-io/pinecone-claude-code-plugin/claude-mdgit clone --depth 1 https://github.com/pinecone-io/pinecone-claude-code-pluginWhat 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 | $0.00680 | $0.00680 |
| Opus 5 | $0.00340 | $0.00340 |
| Sonnet 5 | $0.00136 | $0.00136 |
| Haiku 4.5 | $0.00068 | $0.00068 |
Grade A, and why
pinecone-claude-code-plugin 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 2d 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 — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Claude Code Plugin — Skill Conventions
This file tells Claude how to adapt incoming skills from the base Pinecone skills repo for use in this Claude Code plugin.
Context
This is the Pinecone Claude Code plugin. Skills here are invoked by Claude Code users via the plugin system. The base skills repo is environment-agnostic — this file describes what to change when adapting those skills for Claude Code specifically.
Skill Naming
The base repo uses hyphenated names like pinecone-quickstart. Claude Code uses colon-namespaced names like pinecone:quickstart.
Replace the pinecone- prefix with pinecone: in the name frontmatter field and in any cross-references between skills throughout the content.
Rename skill directories to strip the pinecone- prefix — the plugin already namespaces with pinecone:, so the directory should just be the skill name (e.g., assistant/, quickstart/, cli/).
AskUserQuestion
The base skills use plain prose like "ask the user which path they want" to stay generic. In Claude Code, replace these with the AskUserQuestion tool where it improves the interaction — multi-choice selections, confirmations before running scripts, etc.
Environment Variables
The base skills include .env file fallback instructions for IDEs that don't inherit shell variables. Claude Code runs in the user's terminal session and inherits the shell environment, so:
- Keep the primary
export PINECONE_API_KEY="your-key"instruction - Remove or de-emphasize the
.envfallback — condense to a one-liner note at most
allowed-tools
The base skills don't include an allowed-tools frontmatter field. Add it when adapting skills for Claude Code.
For skills that invoke other skills, always include Skill in allowed-tools:
allowed-tools: Skill, Bash, Read
If a skill only calls specific sub-skills, scope it:
allowed-tools: Skill(pinecone:assistant *), Bash, Read
MCP
You can safely assume the Pinecone MCP will be installed if a user installs the Claude Code Plugin. Check the .mcp.json file if in doubt.
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.
- 2d ago First seen · 77 lines · 680 tokens per session scan A a277db982a85
pinecone-claude-code-plugin CLAUDE.md is an instructions file published in the GitHub repository pinecone-io/pinecone-claude-code-plugin (68 stars, last pushed 18d ago), licensed MIT. It adds 680 tokens to every session, about $0.0034 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.
Other instructions, from other repositories
skills CLAUDE.md
Instructions for pinecone-io/skills, covering skills repo — authoring guidelines, generalization rules and linting.
gemini-cli-extension GEMINI.md
Instructions for pinecone-io/gemini-cli-extension, covering pinecone extension for gemini cli, available agent skills, key concepts & setup and available mcp tools.
ragflow copilot-instructions.md
Copilot instructions for infiniflow/ragflow, covering project instructions for copilot, how to run (minimum), project layout (what matters) and conventions.
chunkhound AGENTS.md
AGENTS.md instructions for chunkhound/chunkhound, covering chunkhound llm context, projectidentity, modificationrules, keycommands and development.
moss AGENTS.md
Instructions for usemoss/moss, covering agents.md, what this repo is, repository layout, integrations & cookbooks and framework cookbooks (examples/cookbook/).
memex AGENTS.md
Instructions for nicosuave/memex, covering agents and before committing.