rango AGENTS.md

A set of project instructions for Rango, an AI copilot layer for Chameleon Ultra. It documents the project layout, installation, runtime paths, verification, and development commands.

In plain words
What is it for?
Working on Rango’s Python library and Kimi Code plugin, including device bridges, card analysis, dictionaries, tests, installation, and runtime setup.
Why use it?
It gives coding agents the project-specific information needed to find files, install the plugin, run it, and verify changes correctly.

Instructions file for CodexOpenCode

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/fedroraddict/rango/agents-md
Clone the repo
git clone --depth 1 https://github.com/fedroraddict/rango

Made for: Codex, OpenCode.

Per session 984 This file is loaded in full into every session.
When invoked 984 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 $0.00984 $0.00984
Opus 5 $0.00492 $0.00492
Sonnet 5 $0.00197 $0.00197
Haiku 4.5 $0.00098 $0.00098

Measured yesterday against content hash ac6f9f18c99c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

rango 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 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.

AGENTS.md · 66 lines

How it starts

The opening of the file, as written. The whole thing — 66 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Rango workspace notes

Rango = the AI copilot layer for the Chameleon Ultra: ai_shell/ (Python wrapper library) + plugin/ (Kimi Code plugin). The upstream Chameleon Ultra CLI is not vendored — scripts/install-cli.sh fetches it (git clone, codeload tarball fallback); at runtime it is located via $CHAMELEON_SOFTWARE/script, ../ChameleonUltra/software/script, then ~/Workspace/chameleonUltra/software/script (see ai_shell/bridge.py).

Layout

  • ai_shell/ — bridge (in-process exec_cmd + stdout/stderr capture), ops.py (state snapshot, slot picking, verified card-load sequence), library.py (host card library), dictionaries.py (key dicts), analyze.py (offline dump analyzer, Flipper .nfc input, known-system key fingerprints, SAK/NTAG type ID), Kimi client, enhanced REPL, selfcheck.py + test_analyze.py.
  • plugin/ — Kimi Code plugin (manifest + MCP server + skill + agents/card-analyst.md). Install with /plugins install <this repo>/plugin then /reload; reinstall after edits (installs are copied to ~/.kimi-code/plugins/managed/). The MCP server runs on this repo's .venv via plugin/run-mcp.sh, which locates the repo via $RANGO_ROOTplugin/.rango-root (written by scripts/setup-plugin.sh) → in-place layout → ~/Workspace/rango, ~/rango.
  • scripts/install-cli.sh (upstream CLI fetcher), setup-plugin.sh (writes plugin/.rango-root; re-run it if the repo moves, then reinstall the plugin).
  • adapters/ — per-harness config/snippets for running the MCP server, skill, and card-analyst agent from Claude Code, Codex CLI, and OpenCode (the plugin manifest is Kimi-specific; the MCP server is not).
  • .github/workflows/ci.yml — CI: ruff, tests, selfcheck (fetches the upstream CLI via scripts/install-cli.sh on the runner).

Verify changes

uv run ruff check ai_shell/ plugin/mcp/
uv run python -m ai_shell.selfcheck     # cited commands exist upstream + gate semantics + slot-pick guards + simulated /plugins install boot
uv run python -m ai_shell.test_analyze  # dump-analyzer regression tests
echo -e "hw version\nexit" | uv run python -m ai_shell   # offline smoke test

Read the full file on GitHub · 66 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 · 66 lines · 984 tokens per session scan A ac6f9f18c99c

Subscribe to this mod's changes

rango AGENTS.md is an instructions file published in the GitHub repository fedroraddict/rango (0 stars, last pushed 9d ago), licensed MIT. It adds 984 tokens to every session, about $0.0049 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.