aios GEMINI.md

A set of project instructions for an AI coding agent working on the AIOS repository. It explains when to use code-relationship tools, how to plan changes, and how to limit safety checks.

In plain words
What is it for?
Guiding repository orientation, planning, impact checks, code searches, and post-edit verification in AIOS projects.
Why use it?
It gives the agent a consistent way to understand the repository and assess risky changes without repeatedly exploring the same information or making unnecessary tool calls.

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/rexleimo/aios/gemini-md
Clone the repo
git clone --depth 1 https://github.com/rexleimo/aios

Made for: Gemini CLI.

Per session 1,138 This file is loaded in full into every session.
When invoked 1,138 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 $0.01138 $0.01138
Opus 5 $0.00569 $0.00569
Sonnet 5 $0.00228 $0.00228
Haiku 4.5 $0.00114 $0.00114

Measured 2d ago against content hash dda0986d80a8, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

aios 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 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.

GEMINI.md · 94 lines

How it starts

The opening of the file, as written. The whole thing — 94 lines — stays where its author put it; the contents beside it link to each section on GitHub.

MCP Tools: code-review-graph

This project exposes a structural knowledge graph via the code-review-graph MCP. Use it only when structural relationships materially affect the current decision; do not turn routine work into a graph-tool loop.

Bounded checkpoints

  • Initial orientation: call get_minimal_context(task="...") at most once when repository structure is not already clear.
  • Before a risky or multi-file change: use get_impact_radius(detail_level="minimal"); call query_graph(pattern="tests_for", target="...") only for the concrete target being changed.
  • After edits: call detect_changes(detail_level="minimal") once when the graph was used or the change has meaningful blast radius.
  • Before submitting: use get_affected_flows() or get_suggested_questions() only if unresolved structural risk remains.
  • Finding code: prefer semantic_search_nodes when semantic graph search is likely to beat a direct repository search.
  • Budget: no more than three graph calls per work item. Treat next_tool_suggestions as optional hints and never follow them recursively.

Planning context proposals

When an active structured-plan task has implementation targets, call AIOS MCP aios_plan_task with action="propose_context", the task id, and workspace-relative targets when the task has none. The tool derives target, caller, callee, and test candidates from codemap, but it does not modify the active plan. Present the candidate refs to a human. An explicit human-controlled CLI confirmation with aios plan task <id> --confirm-context-candidates (optionally repeated --candidate-ref <ref>) activates selected refs for orchestration; it is a process boundary, not an identity/authentication boundary. Do not claim context will be delivered before that command succeeds.

AIOS native enhancements are active in this repository. This is the shared workflow core for every supported coding client.

Read the full file on GitHub · 94 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 · 94 lines · 1,138 tokens per session scan A dda0986d80a8

Subscribe to this mod's changes

aios GEMINI.md is an instructions file published in the GitHub repository rexleimo/aios (52 stars, last pushed 3d ago), licensed MIT. It adds 1,138 tokens to every session, about $0.0057 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.