vector-plugin GEMINI.md

Instructions for using Vector, a red-team scanning service for testing AI agents, from Gemini CLI.

In plain words
What is it for?
They help start scans, retrieve reports, manage agent profiles, integrate scanning into a code repository, and fix failed findings.
Why use it?
They explain when to fetch scan data and when to change code in response to security findings.

Instructions file for Gemini CLI

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/pharosone/vector-plugin/gemini-md
Clone the repo
git clone --depth 1 https://github.com/pharosone/vector-plugin

Made for: Gemini CLI.

Per session 710 This file is loaded in full into every session.
When invoked 710 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 $0.00710 $0.00710
Opus 5 $0.00355 $0.00355
Sonnet 5 $0.00142 $0.00142
Haiku 4.5 $0.00071 $0.00071

Measured yesterday against content hash 738879062b57, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

vector-plugin GEMINI.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 yesterday.

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.

GEMINI.md · 29 lines

How it starts

The opening of the file, as written. The whole thing — 29 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Vector context for Gemini CLI

This Gemini CLI session has access to Vector, a red-team scanning service for LLM agents published by Pharos One. The Vector MCP server is bridged in through mcp-remote and exposes tools that mirror the public REST API (create_session, get_session, list_attacks, submit_results, wait_for_report, get_report, plus the agents.* CRUD).

When to use MCP tools vs skills

  • Use MCP tools when the user wants to do something with Vector data right now from this session: start a scan, fetch the latest report, list attacks for a session, save an agent profile. These return JSON; you summarize.
  • Use a skill when the user wants to change their codebase in response to Vector output: integrate scanning into the repo / CI (integrate), fix a single FAIL (harden-from-finding), batch-fix many FAILs (batch-fix-findings), or generate an AgentContext JSON for the cabinet (create-agent-context).

Skills in Gemini CLI are not invoked through slash-commands — they activate when the user's plain-language request matches the skill's description. So "integrate Vector into this repo" maps to the integrate skill; "help me fix these red-team findings" maps to batch-fix-findings.

Defaults

  • Cabinet (browser UI): https://vector.pharosone.ai.
  • REST API base URL: https://vector-api.pharosone.ai.
  • MCP endpoint: https://vector-api.pharosone.ai/api/v1/mcp/ (trailing slash matters — the server 307-redirects the bare /mcp; overridable via the VECTOR_MCP_URL env var for private deployments).
  • Auth split (important):
    • MCP tools (this session) use OAuth 2.1 via Clerk — browser flow on first call, tokens cached by mcp-remote. No key for you to handle.
    • REST API (the CI runner that the integrate skill writes into the user's repo) uses Authorization: Bearer ak_..., a long-lived Clerk API Key minted by the user in the cabinet Settings → API keys page. Do NOT ask the user to paste the API key value into this chat — the value belongs in their secret manager and a gitignored .env, never in your context window.

Read the full file on GitHub · 29 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. yesterday First seen · 29 lines · 710 tokens per session scan A 738879062b57

Subscribe to this mod's changes

vector-plugin GEMINI.md is an instructions file published in the GitHub repository pharosone/vector-plugin (1 stars, last pushed 3mo ago), licensed MIT. It adds 710 tokens to every session, about $0.0036 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.

Related

Other instructions, from other repositories

midplane AGENTS.md

Instructions for midplaneai/midplane, covering agent operating notes — midplane-cloud, monorepo layout (one codebase, two deployables), design system, client-component imports from @midplane-cloud/db and server actions: return state, don't throw, for user input.

midplaneai/midplane · 2,280 tokens

midplane CLAUDE.md

Instructions for midplaneai/midplane, covering claude code — midplane-cloud and skill routing (claude code).

midplaneai/midplane · 263 tokens

niuma-engine CLAUDE.md

Instructions for Destined-at-Dawn/niuma-engine, covering startup sequence (every conversation -- mandatory), project rules, r21 -- three-layer file protection (critical), think before act and no blind overwrite (critical).

Destined-at-Dawn/niuma-engine · 1,512 tokens

agentbreak CLAUDE.md

Instructions for mnvsk97/agentbreak, covering claude code — agentbreak, what this repo is, quick commands, skill and verification.

mnvsk97/agentbreak · 155 tokens

pattern8 AGENTS.md

Instructions for Aquifer-sea/pattern8, covering agents.md — p8 global instructions (all agents must follow), pattern 8 (p8) — agent behavior constraints, core constraints, five patterns (cannot be skipped) and 1. pipeline.

Aquifer-sea/pattern8 · 414 tokens

awesome-copilot-id AGENTS.md

Instructions for GulajavaMinistudio/awesome-copilot-id, covering communication, explanation and documentation, markdown formatting, user communication style and workflow & methodology.

GulajavaMinistudio/awesome-copilot-id · 7,686 tokens