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/jujitae/invara/agents-mdgit clone --depth 1 https://github.com/Jujitae/invaraWrote 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/jujitae/invara/agents-md)<a href="https://agentmods.dev/instructions/jujitae/invara/agents-md"><img src="https://agentmods.dev/badge/instructions/jujitae/invara/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 | $0.00554 | $0.00554 |
| Opus 5 | $0.00277 | $0.00277 |
| Sonnet 5 | $0.00111 | $0.00111 |
| Haiku 4.5 | $0.00055 | $0.00055 |
Grade A, and why
invara 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 — 56 lines — stays where its author put it; the contents beside it link to each section on GitHub.
INVARA local execution handbook
Product boundary
INVARA independently evaluates declared constraints from observable file bytes,
command results, and recorded provenance. It does not prove general software
correctness, authorize work, or accept an agent's claim that work is done.
BLOCK and UNVERIFIABLE are valid fail-closed outcomes, not conditions to
weaken away.
Local architecture and canonical sources
src/invara/: contract sealing, observation, verdict, replay, and hash-chain implementation.tests/test_invara.py: refusal-first executable specification.README.md: user-facing product contract and first-verdict flow.PROVENANCE.md,PRIVACY.md, andSECURITY.md: provenance, local-only data, and threat-model commitments.plugin/: bundled MCP/stdio surface; its Python package and license files must stay byte-aligned withsrc/invara/and the root license files.- Company orchestration, portfolio priority, role routing, and main integration
authority live in
migaryos-hq; this repository does not restate them.
Invariants and known traps
- Observable evidence and deterministic verdicts are authoritative; self-report is not completion evidence. An agent may request a verdict but cannot give itself one.
- Preserve sealed-contract, protected-path, evidence, hash-chain, and replay
semantics. Do not turn a refusal, broken check, or missing executable into
green; retain fail-closed
UNVERIFIABLE/BLOCKbehavior. - The buyer is a non-coder: the MCP/stdio entry point and usable first-verdict experience are product surfaces, not optional demos.
- Keep the zero-runtime-dependency and local-only/no-network claims true. Contracts run user-named commands with user permissions and can store their command strings; never put secrets in a contract.
- Windows encoding, Store Python aliases, CRLF digest changes, and plugin-copy drift are tested product behavior. Do not paper over them with platform- specific bypasses.
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 · 56 lines · 554 tokens per session scan A 323554663581
invara AGENTS.md is an instructions file published in the GitHub repository Jujitae/invara (1 stars, last pushed 2d ago), licensed Apache-2.0. It adds 554 tokens to every session, about $0.0028 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 2026-09-04.
Other instructions, from other repositories
xchecker CLAUDE.md
Claude Code instructions for EffortlessMetrics/xchecker, covering claude.md, project overview, workspace layout, build & test commands and build.
xchecker AGENTS.md
AGENTS.md instructions for EffortlessMetrics/xchecker, covering claude.md, project overview, workspace layout, build & test commands and build.
std-env AGENTS.md
AGENTS.md instructions for unjs/std-env, covering agents, project overview, maintenance, scripts and source structure.
flyto-core AGENTS.md
AGENTS.md instructions for flytohub/flyto-core, covering flyto-core agent rules, flyto2 project memory contract, flyto2 frontend quality gate and repo notes.
flyto-core CLAUDE.md
Claude Code instructions for flytohub/flyto-core, covering claude notes, cross-agent handoff and shared code intelligence.
teamcity-mcp CLAUDE.md
Claude Code instructions for Daghis/teamcity-mcp, covering claude code instructions for teamcity-mcp, project overview, development commands, testing and unit tests.