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.
npx agentmods add instructions/sergkam/flycrys/agents-mdgit clone --depth 1 https://github.com/SergKam/FlyCrysWrote 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/instructions/sergkam/flycrys/agents-md)<a href="https://agentmods.dev/instructions/sergkam/flycrys/agents-md"><img src="https://agentmods.dev/badge/instructions/sergkam/flycrys/agents-md.svg" alt="Measured on agentmods" height="20"></a>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 | $0.03102 | $0.03102 |
| Opus 5 | $0.01551 | $0.01551 |
| Sonnet 5 | $0.00620 | $0.00620 |
| Haiku 4.5 | $0.00310 | $0.00310 |
Grade A, and why
FlyCrys 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 4d 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.
How it starts
The opening of the file, as written. The whole thing — 252 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md - FlyCrys
Instructions for AI agents working on this codebase.
Rules
-
NEVER use deprecated GTK4 APIs. No TreeView, TreeStore, CellRenderer*, ListStore (deprecated one). Use modern replacements: ListView, ColumnView, TreeListModel, SignalListItemFactory, TreeExpander, gio::ListStore. No
#[allow(deprecated)]annotations. -
ALWAYS check latest GTK4 API docs before using any GTK API. Do NOT rely on memory or training data — GTK4 APIs change between versions and many are deprecated. Use Context7 (
resolve-library-id+query-docsforgtk4-rs) or fetch the official docs at https://docs.rs/gtk4 and https://docs.gtk.org/gtk4/ to verify that the API you're about to use is current and not deprecated. When in doubt, look it up. -
Handle errors properly — show the reason to the user. Never silently swallow errors. Follow these rules:
In UI code: Show meaningful error messages to the user. Use
gtk::AlertDialog(NOT the deprecatedMessageDialog):// Correct pattern — AlertDialog with detail let dialog = gtk::AlertDialog::builder() .message("Operation failed") .detail(&format!("Could not open file: {err}")) .build(); dialog.show(Some(&window));For errors inside the agent panel, use
agent_widgets::create_system_message()to display the error inline with the reason.In service code: Return
Result<T, String>(or a proper error type) with a descriptive message that includes the underlying cause. Never discard thewhy:// Bad — reason is lost fs::read_to_string(path).ok(); command.spawn().ok(); // Good — reason is propagated fs::read_to_string(path).map_err(|e| format!("Failed to read {}: {e}", path.display()))?;Specifically:
eprintln!()is acceptable for debug/logging but is NOT a substitute for user-visible feedback.unwrap()is only acceptable on infallible operations (e.g., GTK widget downcasts inside factory callbacks where the type is guaranteed). Never on I/O, parsing, or network operations..ok()that discards a result is a code smell — justify it with a comment or handle the error- For async dialog callbacks (
FileDialog,AlertDialog), check forGtkDialogError::Dismissed— that's the user choosing to cancel, not an error worth reporting
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.
- 4d ago First seen · 252 lines · 3,102 tokens per session scan A 6f01fa415497
FlyCrys AGENTS.md is an instructions file published in the GitHub repository SergKam/FlyCrys (31 stars, last pushed 2mo ago), licensed MIT. It adds 3,102 tokens to every session, about $0.0155 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.
Other instructions, from other repositories
freya AGENTS.md
AGENTS.md instructions for marc2332/freya, covering agents, commands, architecture, rust and general instructions.
codex-manager AGENTS.md
AGENTS.md instructions for ewsun22/codex-manager, covering codex manager 项目指令, 产品边界, 隐私与安全, 工程规则 and 开源文档与发布流程.
leantoken AGENTS.md
Instructions for morluto/leantoken, covering repository guidance, architecture, development, change-specific validation and contributions.
swarm AGENTS.md
Instructions for penberg/swarm, covering agents and coding style.
sessions-chronicle AGENTS.md
Instructions for supermaciz/sessions-chronicle, covering repository guidelines, where to look first, terminology, project structure & module organization and fast dev loop.
r-code AGENTS.md
Instructions for foritin/r-code, covering agents.md — r-code 仓库开发规约, 工具使用速查(避免低级报错), 通则 and 本仓库技术栈.