agentmods.dev

vector-mcp AGENTS.md

Repository-specific instructions for vector-mcp, a server that lets agents search and manage collections of connected information.

In plain words
What is it for?
Use them when changing the vector-mcp server, its agent entry point, search or collection tools, provider adapters, skills, or data model.
Why use it?
They define the architecture and non-negotiable rules for validation, privacy, credentials, storage, and optional backends.

Instructions file for CodexOpenCode

15 4d ago A scan 2,948 tokens original MIT
View the source file Knuckles-Team/vector-mcp
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/knuckles-team/vector-mcp/agents-md
Clone the repo
git clone --depth 1 https://github.com/Knuckles-Team/vector-mcp

Made for: Codex, OpenCode.

Per session 2,948 This file is loaded in full into every session.
When invoked 2,948 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 $0.02948 $0.02948
Opus 5 $0.01474 $0.01474
Sonnet 5 $0.00590 $0.00590
Haiku 4.5 $0.00295 $0.00295

Measured 2d ago against content hash fa55a5828a3c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

vector-mcp AGENTS.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.

AGENTS.md · 216 lines

How it starts

The opening of the file, as written. The whole thing — 216 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md

Scope

This repository provides the action-routed vector-mcp server, optional vector provider adapters, an agent entry point, packaged skills/prompts, a vector ontology, and a read-only collection-inventory connector preset.

  • Language/Version: Python 3.11+
  • Core Libraries: agent-utilities, fastmcp, pydantic-ai
  • Key principles: Functional patterns, Pydantic for data validation, asynchronous tool execution.

Current architecture

MCP client
  -> vector_collection_management / vector_search
  -> validation, privacy, and backend policy
  -> verified GraphSession boundary
  -> configured provider

Epistemic-graph is the native default. Optional providers are isolated behind extras and imported only when selected.

Non-negotiable design rules

  • Do not add legacy paths, fallback implementations, deprecated aliases, compatibility shims, or backward-compatibility branches. Migrate the current contract directly.
  • Do not accept credentials or local database paths as MCP tool arguments.
  • Do not persist endpoints, credentials, personal identity, raw content, hostnames, or local filesystem paths in source, skills, docs, traces, reports, fixtures, or generated evidence.
  • Resolve runtime values through AgentConfig, environment variables, and supported secret references.
  • Resolve TLS through the shared transport profile. Do not add certificate-verification bypasses.
  • Keep document ingestion confined to the configured root and relative caller-selected paths.
  • Keep optional dependencies lazy; importing vector_mcp must not start services or require every provider SDK.
  • Never fabricate connector signatures, live test evidence, trace evidence, or safe dependency versions.

Source layout

  • vector_mcp/mcp_server.py: condensed MCP registration and request boundary
  • vector_mcp/vector_api.py: verified session, privacy, secret, TLS, and provider boundary
  • vector_mcp/vectordb/: optional provider adapters
  • vector_mcp/backend_policy.py: canonical backend and exposure policy
  • vector_mcp/document_inputs.py: bounded root-confined ingestion inputs
  • vector_mcp/doctor.py: privacy-safe readiness output
  • vector_mcp/skills/, prompts/, ontology/, connectors/: packaged extensions
  • tests/: unit and explicitly configured integration tests
  • docs/: operator documentation

Read the full file on GitHub · 216 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. 2d ago First seen · 216 lines · 2,948 tokens per session scan A fa55a5828a3c

Subscribe to this mod's changes

vector-mcp AGENTS.md is an instructions file published in the GitHub repository Knuckles-Team/vector-mcp (15 stars, last pushed 4d ago), licensed MIT. It adds 2,948 tokens to every session, about $0.0147 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.

Data and AIPlanningResearchchromadbcouchbasemcp-servermongodb-atlaspgvector
Related

Other instructions, from other repositories

mempalace CLAUDE.md

Claude Code instructions for MemPalace/mempalace, covering claude.md, the mission, design principles, contributing and setup.

MemPalace/mempalace · 1,683 tokens

mongodb-mcp-server copilot-instructions.md

Copilot instructions for mongodb-js/mongodb-mcp-server, covering project overview, folder structure, libraries and frameworks, coding standards and architectural guidelines and best practices.

mongodb-js/mongodb-mcp-server · 1,576 tokens

eShopLite copilot-instructions.md

Instructions for Azure-Samples/eShopLite, covering eshoplite copilot instructions, project overview, architecture & patterns, critical workflows and project-specific conventions.

Azure-Samples/eShopLite · 1,027 tokens

pgmnemo AGENTS.md

Instructions for pgmnemo/pgmnemo, covering pgmnemo — agent integration guide, 1. what pgmnemo is — and the problem it solves, the problem, architectural consequences and 2. when to adopt / when not to.

pgmnemo/pgmnemo · 8,496 tokens

mempalace AGENTS.md

AGENTS.md instructions for MemPalace/mempalace, a project described as: The best-benchmarked open-source AI memory system. And it's free.

MemPalace/mempalace · 4 tokens

cerefox CLAUDE.md

Instructions for fstamatelopoulos/cerefox, covering cerefox - project guide, what is this, tech stack, project structure and development conventions.

fstamatelopoulos/cerefox · 10,388 tokens