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/rx6ru/strata/agents-mdgit clone --depth 1 https://github.com/rx6ru/strataWhat 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 | $0.00429 | $0.00429 |
| Opus 5 | $0.00215 | $0.00215 |
| Sonnet 5 | $0.00086 | $0.00086 |
| Haiku 4.5 | $0.00043 | $0.00043 |
Grade A, and why
strata 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.
How it starts
The opening of the file, as written. The whole thing — 40 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Operating Contract
This repository builds a generic Strata service for agents. Keep it public-safe, portable, and secret-free.
Non-negotiable rules
- Never commit secrets, credentials, tokens, private keys, cookies, environment values, provider settings, host-specific SSH details, cloud instance IDs, private IPs, or local personal paths.
- Do not hardcode deployment-specific values in code, docs, examples, tests, or the skill. Use placeholders and require the runtime user or agent to provide values.
- Treat canonical storage as append-only. Do not add ordinary overwrite or delete paths.
- Treat derived memory or recall engines as rebuildable caches, not authoritative truth.
- Keep the server LLM-free. The caller's active agent may use its own model outside the service.
- Preserve portability: a complete export/restore round trip must keep the root manifest consistent.
Before writing memory-related changes
- Read
policy/*.mdand the affected code path. - Add or update tests before changing behavior when coverage is missing.
- Verify secret rejection, policy acknowledgement, append-only semantics, and namespace scoping for the changed path.
- Run the smallest relevant test first, then a broader test pass before completion.
Documentation rules
- Examples must be generic and executable from a public checkout.
- Use placeholders such as
<gateway-url>,<namespace>, and<ssh-target-or-alias>for runtime values. - If a command requires optional infrastructure, state the prerequisite instead of implying it works everywhere.
- Do not mention private users, machines, projects, repositories, IP addresses, tokens, or instance identifiers.
Release checklist
python -m unittest discover -s tests -vpython -m compileall -q strata tests tools- Docker/Compose configuration check when Docker is available
- skill reference integrity check
- secret scan over all tracked files
- internal-reference scan over all tracked files
- export/restore smoke test
- public README and docs reviewed for unsupported claims
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.
- yesterday First seen · 40 lines · 429 tokens per session scan A 6921ab63a666
strata AGENTS.md is an instructions file published in the GitHub repository rx6ru/strata (0 stars, last pushed 6d ago), licensed MIT. It adds 429 tokens to every session, about $0.0021 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
Binder AGENTS.md
Instructions for mpazik/Binder, covering binder, tech stack, monorepo structure, testing and development.
trip2g CLAUDE.md
Instructions for trip2g/trip2g, covering trip2g, rules, build commands, critical: default template pipeline and key patterns.
trip2g AGENTS.md
Instructions for trip2g/trip2g: Read CLAUDE.md before starting work in this repository. Treat its rules, patterns, build commands, and documentation guidance as mandatory project instructions.
kb AGENTS.md
Instructions for hraness/kb, a project described as: A knowledge base for coding agents, built from Markdown, backlinks, semantic search, and Git context.
artel CLAUDE.md
Instructions for NicolasPrimeau/artel, covering artel, what it is, stack, layout and core primitives.
signetai AGENTS.md
AGENTS.md instructions for Signet-AI/signetai, covering working in signet, start with the checkout, find the right context, durable data contracts and verification.