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 skills add PostOakLabs/ainumbers --skill two-hosts-one-hashgit clone --depth 1 https://github.com/PostOakLabs/ainumbersWrote 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/skills/postoaklabs/ainumbers/two-hosts-one-hash)<a href="https://agentmods.dev/skills/postoaklabs/ainumbers/two-hosts-one-hash"><img src="https://agentmods.dev/badge/skills/postoaklabs/ainumbers/two-hosts-one-hash/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/skills/postoaklabs/ainumbers/two-hosts-one-hash"><img src="https://agentmods.dev/badge/skills/postoaklabs/ainumbers/two-hosts-one-hash.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00048 | $0.00549 |
| Opus 5.5 | $0.00019 | $0.00220 |
| Sonnet 5.5 | $0.00010 | $0.00110 |
| Haiku 4.5 | $0.00005 | $0.00055 |
Grade A, and why
two-hosts-one-hash 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 15d 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.
What it actually says
Two hosts return one hash
Run the same tool from two different agents and prove they agree on the execution hash.
Prompt
Prove that two independent agent hosts get byte-identical results from the same tool.
- Pick one node tool (any page in /chaingraph/ works). Call find_tool to confirm its exact mcp_name and inputSchema.
- Fix one synthetic input once, write it down, and do not change it for the rest of the run.
- Host A: run the tool through the hosted MCP worker with that input. Record execution_hash.
- Host B: a different agent (Claude, OpenClaw, Goose, anything that speaks MCP) runs the same tool with the same input through the same worker. Record execution_hash.
- Compare the two hashes character by character. If they differ, the input was not identical: diff both recorded inputs and re-run until you have either two equal hashes or a proven input difference.
- Call verify_execution_hash yourself over {policy_parameters, output_payload} to confirm the hash is the deterministic function both hosts should compute.
- build_session_receipt over the comparison and give me the ledger link.
- Write the note for your team's runbook: which hosts agreed, the hash, and the ledger link, so the next person repeats this with their own agents. If a named tool is not in your tool list, call call_tool with { "name": "", "arguments": { … } }. It runs the same validation and returns the same receipt.
How to run
Connect an MCP-capable host to https://mcp.ainumbers.co/mcp, then give the prompt above to the assistant.
Use synthetic inputs only. Never paste personal or production data.
Verify what came back
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.
- 15d ago Changed 150de02fa626
- 19d ago First seen · 40 lines · 48 tokens per session scan A 920cd9441777
two-hosts-one-hash is a skill published in the GitHub repository PostOakLabs/ainumbers (11 stars, last pushed yesterday), licensed MIT. It adds 48 tokens to every session and 549 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-09-27.
Other skills, from other repositories
poc-validator
Proof-of-Concept (PoC) Validation Skill for security exploit verification. Use when validating exploitability of vulnerabilities, generating tailored payloads, executing exploits in sandboxed environments, or verifying successful exploitation (e.g., SQL injection, CVE exploitation, command injection). Triggers on…
tdd-agent
Autonomous test-driven development agent that writes code to make tests pass.
bash-testing
Adds or updates bats-core, shunit2, or plain-script test coverage for a Bash change by first discovering the repository's existing test layout and conventions, matching them rather than imposing a new framework, deciding whether the change requires a test at all, making a script testable by separating logic into…
mutation-testing
Run and READ a mutation pass on this repo (StrykerJS): when a pass is due, which runner to use for which package, how to batch it, and how to judge the survivors — kill, simplify the production, or name an honest equivalent. Use when about to mutate a file, when a run has just produced survivors, when a score has to…
hr-qa
Help HR managers, recruiters, and talent acquisition teams understand Quality Assurance (QA), Software Testing, Test Automation, and modern Quality Engineering concepts. Use when asked to explain QA testing, screen QA candidates, understand automation testing, compare manual QA and SDET, evaluate QA skills, create QA…
accessibility-audit
Use when wCAG compliance audit — semantic HTML, ARIA, keyboard navigation, color contrast, and screen reader testing. Use when working with accessibility audit.