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/vbkotecha/agentservices-api/agents-mdgit clone --depth 1 https://github.com/vbkotecha/agentservices-apiWrote 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/vbkotecha/agentservices-api/agents-md)<a href="https://agentmods.dev/instructions/vbkotecha/agentservices-api/agents-md"><img src="https://agentmods.dev/badge/instructions/vbkotecha/agentservices-api/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 | $0.01164 | $0.01164 |
| Opus 5 | $0.00582 | $0.00582 |
| Sonnet 5 | $0.00233 | $0.00233 |
| Haiku 4.5 | $0.00116 | $0.00116 |
Grade A, and why
agentservices-api 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 today.
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 — 98 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent instructions — AgentServices API
Locked conventions for cloud agents and autonomous contributors. Follow on every trade-related change.
WAR TIME trade API
Path model: venue-namespaced doors first.
| Phase | Path | Status |
|---|---|---|
| Now | /v1/trade/hyperliquid/... |
Shipped |
| Next | /v1/trade/{venue}/... |
Add venues under this prefix |
| Later | /v1/trade/execute |
Venue-neutral router (not built yet) |
| Never | /v1/trade/router |
Do not create or alias |
Hyperliquid (first venue)
Base: /v1/trade/hyperliquid
| Endpoint | Method | Notes |
|---|---|---|
/order |
POST | Forward agent-signed order |
/cancel |
POST | Forward agent-signed cancel |
/order |
GET | Status (query: user, oid) |
/order/{id} |
GET | Status by path param |
/policy |
GET / PUT | Execution leash |
/paper/order |
POST | Paper training |
/eval/order |
POST | Policy pass/fail eval |
/bootstrap |
GET | approveAgent signing docs |
Request body: include market_type — spot, perp, or future (validated). Hyperliquid implements perp + spot today; unsupported types return machine-readable market_type_not_supported (422).
MCP tools: trade_hyperliquid_* (primary). Legacy hl_* names are aliases only.
Trade invariants (all venues)
- Free policy leash — max notional, coin allowlist, kill switch. No x402 on execution paths.
- No venue key custody — agents sign locally (
approveAgent+ local agent wallet). AgentServices never collects venue API keys or user private keys. - Door + forward — policy check then forward signed payload. Not a smarter router; not a better fill.
Dual trade door — execution vs monetization
Two doors, one product. Do not collapse them.
| Door | What | Monetization |
|---|---|---|
| Execution | /v1/trade/{venue}/... (HL today), later /v1/trade/execute |
Always free of x402 paywall. Policy leash only. |
| Data / discovery | prices, indicators, search, synthesis, MCP paid tools | x402 per-request SKUs (unchanged). |
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.
- today First seen · 98 lines · 1,164 tokens per session scan A f8dba94e49b3
agentservices-api AGENTS.md is an instructions file published in the GitHub repository vbkotecha/agentservices-api (1 stars, last pushed today), licensed Apache-2.0. It adds 1,164 tokens to every session, about $0.0058 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
bharatpay-mcp CLAUDE.md
Claude Code instructions for angelina10504/bharatpay-mcp, covering claude.md, what this is, layout, engineering standards and 1. every tool has a pure-function offline validator.
mcp-server AGENTS.md
AGENTS.md instructions for NannyKeeper/mcp-server, covering nannykeeper mcp server — agent instructions, purpose, tools available, when to use and key facts.
mcp-server CLAUDE.md
Claude Code instructions for NannyKeeper/mcp-server, covering nannykeeper mcp server, tools, setup and context for ai assistants.
fli CLAUDE.md
Instructions for punitarani/fli, covering claude.md, project overview, development commands, core development tasks and install dependencies.
gmgn-skills CLAUDE.md
Claude Code instructions for GMGNAI/gmgn-skills, covering claude.md, critical rule — read this first, project overview, available skills and quick decision guide.
maverick-mcp AGENTS.md
AGENTS.md instructions for wshobson/maverick-mcp, covering repository guidelines, project overview, project structure, documentation map and build, test, and development commands.