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.
git clone --depth 1 https://github.com/ilang-ai/autocodeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/plugins/ilang-ai/autocode/gemini-extension)<a href="https://agentmods.dev/plugins/ilang-ai/autocode/gemini-extension"><img src="https://agentmods.dev/badge/plugins/ilang-ai/autocode/gemini-extension/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/plugins/ilang-ai/autocode/gemini-extension"><img src="https://agentmods.dev/badge/plugins/ilang-ai/autocode/gemini-extension.svg" alt="Reviewed on agentmods" width="80" height="20"></a>Grade A, and why
autocode 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 today.
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.
What it actually says
{
"name": "autocode",
"version": "6.2.1",
"description": "You say it, AutoCode ships it. 11 skills. See a product you like → sourced blueprint → MVP → deployed URL. I-Lang v5.0 powered.",
"contextFile": "README.md"
}
What ships with it
5 files beside gemini-extension.json in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- today Changed 324ab831dece
- yesterday Changed 464ce2d48110
- 2d ago Changed 7be7d4070371
- 1mo ago First seen · 7 lines scan A 8149562ce46c
autocode is a plugin published in the GitHub repository ilang-ai/autocode (103 stars, last pushed yesterday), licensed MIT. Its token cost is not measured: this kind of file is read by the harness, not the model. 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 2026-09-14.
Other plugins, from other repositories
ast-grep
Guide for writing ast-grep rules to perform structural code search and analysis. Use when users need to search codebases using Abstract Syntax Tree (AST) patterns, find specific code structures, or perform complex code queries that go beyond simple text search.
claude-mem-grok-bot
Persistent memory for Grok Bot. No host hooks: transcript watcher plus MCP. Local worker with host-login observer, or remote cmem.ai. Install without Cursor.
jev-guardrails
Screens every prompt going into the conversation and every reply coming out of it with TypeSafe's Jev, a System One decision model, the way TypeSafe's guardrails cookbook does: one request per message asks a battery of hazard questions (jailbreak, harm or a crime, a diagnosis or a dosage, self-harm) and scores how…
jev-skill-typeahead
Shows, while you type, the skills and subagents Claude will probably call for your prompt: reads the draft on every edit and lists them in the band above the prompt, ranked by keyword match (English or Spanish). When you pause, Jev (TypeSafe's decision model) decides which one will be called and the band marks it.…
jev-auto-mode
A permission layer driven by a JSON policy: allow, ask or deny every tool call (Bash, edits, web, MCP, subagents, skills), every slash command or skill you type, and every skill preloaded into a subagent. Rules decide first (deny > ask > allow, compound Bash commands split); what no rule covers can go to TypeSafe's…
agent-flow
A side pane that draws the session's agent flow live: the main loop's context window, every subagent it spawns as a tree, and for each one the context handed down (its prompt, or the whole parent context for a fork), what it did with it (steps, tools, context size) and the answer it handed back up. Docked beside the…