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/winsznx/flightrules/claude-mdgit clone --depth 1 https://github.com/winsznx/flightrulesWrote 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/winsznx/flightrules/claude-md)<a href="https://agentmods.dev/instructions/winsznx/flightrules/claude-md"><img src="https://agentmods.dev/badge/instructions/winsznx/flightrules/claude-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 | $0.02237 | $0.02237 |
| Opus 5 | $0.01118 | $0.01118 |
| Sonnet 5 | $0.00447 | $0.00447 |
| Haiku 4.5 | $0.00224 | $0.00224 |
Grade A, and why
flightrules CLAUDE.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 3d 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 — 184 lines — stays where its author put it; the contents beside it link to each section on GitHub.
FlightRules — operating contract
This file is the working contract for anyone (human or agent) changing this repository. It is
derived from docs/PRD.md, which remains authoritative. Where this file and the PRD disagree,
the PRD wins and this file is wrong and must be fixed.
design.md at the repository root is authoritative only for colours, typography, spacing,
layout styling, borders, shadows, icon treatment, logos, illustrations, assets, visual motion
and responsive visual rules. It must never change product wording, routes, features, states or
information architecture.
What this product is
FlightRules turns SigNoz traces into deterministic release contracts that catch skipped checks, duplicate side effects, unknown tool paths and behavioural drift before an agent canary reaches production.
The graph diff is the product hero. A chatbot is not the interface.
Non-negotiable rules
- One numbered phase at a time. Do not start a phase whose entry criteria are unmet.
- A phase is complete only when its tests and its runtime validation have both passed.
- Code existing is not evidence a feature works. Runtime evidence is required.
- Never infer an undocumented request or response shape.
- Never invent a SigNoz endpoint, MCP tool, OpenTelemetry attribute, Foundry field, package option, environment variable or CLI flag.
- Verify against installed runtime first, then official versioned docs, then the official repository at the pinned commit. A blog post, tutorial, forum answer or recalled syntax is never the sole basis for an implementation decision.
- Record every external technical decision in
docs/research/source-lock.md. - When documentation and runtime behaviour disagree, the runtime wins for the installed version, and the mismatch is recorded.
- Never silently downgrade a requirement.
- Never replace a required integration with a mock and call the phase complete. Mocks are for unit tests and deterministic offline fixtures only.
- No placeholder
return true, fabricated telemetry, hard-coded dashboard screenshot, invented trace ID, fake MCP response, or manually inserted database row may be used as proof. - No unresolved P0 TODO, FIXME, stub, disabled test, skipped test or commented-out implementation may remain at submission.
- Every MCP write is followed by a read-back that confirms the resource and validates the fields that matter. A successful call is not proof the resource is correct.
- Every generated dashboard panel must return real data after the seeded demo runs.
- Every alert must be driven through a real firing state, and a recovery state where the installed SigNoz version supports it.
- Every trace rule must be tested against both a passing and a violating trace.
mainstays green. Do not merge a phase branch with failing checks.- Focused conventional commits. No co-author trailers.
- Do not rewrite unrelated code while completing a phase.
- No secret in source, logs, test snapshots, screenshots, browser storage or telemetry.
- Never capture chain-of-thought. FlightRules evaluates observable execution structure and safe metadata only.
- Read the surrounding code and its tests before changing an existing file.
- After every phase, update the acceptance matrix, evidence log, ADRs and changelog.
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.
- 3d ago First seen · 184 lines · 2,237 tokens per session scan A e2a7966b14f7
flightrules CLAUDE.md is an instructions file published in the GitHub repository winsznx/flightrules (2 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 2,237 tokens to every session, about $0.0112 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.
Other instructions, from other repositories
openllmetry CLAUDE.md
Instructions for traceloop/openllmetry, covering openllmetry repository guide, repository structure, nx workspace commands, run tests across all packages and run linting across all packages.
clawmetry AGENTS.md
AGENTS.md instructions for vivekchand/clawmetry, covering agents.md — for ai coding agents, quick context, where new code goes (open-core split), the rules that bite and common tasks.
signoz-mcp-server CLAUDE.md
Instructions for SigNoz/signoz-mcp-server, covering claude.md — development conventions, feature planning convention, rules, git & pr and code style.
clawmetry CLAUDE.md
Claude Code instructions for vivekchand/clawmetry, covering claude.md — clawmetry, what is this?, architecture, key files and core.
opentelemetry-agent-skills AGENTS.md
AGENTS.md instructions for ollygarden/opentelemetry-agent-skills, covering agents.md, what this repository is, preferred workflow, running the gates locally and skill constraints.
void-box AGENTS.md
AGENTS.md instructions for the-void-ia/void-box, covering agents.md — voidbox, design proposals: rfc + adr process, code intelligence, commit and pr messages and markdown documentation style.