A set of repository instructions for contributors to the Vault 3000 project. It documents the main command-line and web entry points, setup commands, supported run modes, and configuration details.

It helps developers set up the project, run its local or remote command-line agent, start the API server, use pipeline modes, and avoid documented configuration mistakes.

Contributors otherwise have to discover how the project is structured and run by reading the code or guessing commands.

vault_agent — Instructions file by noxgle

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/noxgle/term_agent/agents-md

Clone the repo

git clone --depth 1 https://github.com/noxgle/term_agent

Made for: Codex, OpenCode.

Per session 607 This file is loaded in full into every session.

When invoked 607 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.

Model	Per session	Once invoked
Fable 5	$0.00607	$0.00607
Opus 5	$0.00303	$0.00303
Sonnet 5	$0.00121	$0.00121
Haiku 4.5	$0.00061	$0.00061

Measured 2d ago against content hash 466604fbb4b9, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

vault_agent 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 · 42 lines

How it starts

The opening of the file, as written. The whole thing — 42 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md

What matters most

term_ag.py is the real CLI entrypoint (local + SSH mode).
VaultAiAgentRunner.py is the core execution pipeline (compact/normal/hybrid behavior).
api/api_server.py + api/api_agent.py are the HTTP path; API runs non-interactively (auto_accept=True, prompt replies forced to "n").
term_api.py only starts uvicorn using API_HOST/API_PORT.

Verified run commands

Setup: python3 -m venv .venv && source .venv/bin/activate && pip install -r requirements.txt
Agent CLI: python term_ag.py
Remote mode: python term_ag.py user@host
Prompt-creator mode: python term_ag.py --prompt (also separate python PromptCreator.py exists)
Pipeline overrides: python term_ag.py --compact|--normal|--hybrid
Force plan from CLI: python term_ag.py --plan
API server: python term_api.py
Docker dev path: docker-compose up -d (SSH on host :2222, API on ${API_PORT:-8000})

Behavior and config quirks (easy to miss)

.env must exist in repo root; term_ag.py exits immediately if missing.
AI_ENGINE supports comma-separated engines; routing is controlled by AI_ENGINE_ROUTE (round-robin or fallback).
Default CLI runtime mode is effectively hybrid unless --compact/--normal overrides are passed.
Ctrl+A in CLI switches from collaborative to automatic mode one-way for current session.
LOG_FILE in .env is written under ./logs/ unless an absolute/explicit path is provided.
API defaults to pipeline_mode="hybrid" when neither pipeline_mode nor compact_mode is supplied.
API remote runs require ssh_password in payload (api/api_agent.py enforces this).

Test reality (current repo state)

No pytest/ruff/mypy config files are present in root.
requirements.txt does not include pytest or pytest-asyncio, even though tests import them.
Current tests/test_mcp_* reference an mcp package path not present in this repository; run tests selectively and expect MCP tests to fail until that module is restored.

Read the full file on GitHub · 42 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.

2d ago First seen · 42 lines · 607 tokens per session scan A 466604fbb4b9

Subscribe to this mod's changes

vault_agent is an instructions file published in the GitHub repository noxgle/term_agent (6 stars, last pushed 3mo ago), licensed MIT. It adds 607 tokens to every session, about $0.0030 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.

Backend and APIs Productivity ai automation cli devtools linux

vault_agent

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

AGENTS.md

What matters most

Verified run commands

Behavior and config quirks (easy to miss)

Test reality (current repo state)

What this file has done since we first saw it

Other instructions, from other repositories

dev3000 AGENTS.md

best-claude-hud AGENTS.md

jenkins-cli CLAUDE.md

mcp-screenshot-server CLAUDE.md

patchloom AGENTS.md

ducklab AGENTS.md