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 skills/observer-protocol/at-directory/skillnpx skills add observer-protocol/at-directory --skill skillgit clone --depth 1 https://github.com/observer-protocol/at-directoryWhat 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 | $0.00054 | $0.01800 |
| Opus 5 | $0.00027 | $0.00900 |
| Sonnet 5 | $0.00011 | $0.00360 |
| Haiku 4.5 | $0.00005 | $0.00180 |
Grade A, and why
agentic-terminal scanned grade A with 1 finding 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl 'https://mcp.agenticterminal.ai/v1/merchants?rail=lightning&category=gift-cards' How it starts
The opening of the file, as written. The whole thing — 118 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AT Directory
AT Directory is the agent commerce discovery surface for Bitcoin- and Tether-aligned payments. It indexes merchants that sell products, services, APIs, or content and accept Lightning, BOLT12, L402, or USDT (any chain), with trust attestations issued through Observer Protocol. It is not in the payment path — you discover merchants here and pay the merchant directly.
1. Connect — three paths, pick one
All three hit the same data and are anonymous + ungated (no credential to discover or transact).
A. Hosted MCP (recommended — lowest friction, no install). Point any MCP-compliant client at the hosted URL:
{ "mcpServers": { "at-directory": { "url": "https://mcp.agenticterminal.ai/mcp" } } }
B. Local MCP (npm). For air-gapped or offline-first setups:
npm install -g @agenticterminal/mcp-server # Node >= 20
at-directory-mcp # stdio MCP server
C. Plain REST (no MCP client needed). For agent runtimes that don't speak MCP (OpenClaw, custom Lightning-native agents) — just HTTP:
curl 'https://mcp.agenticterminal.ai/v1/merchants?rail=lightning&category=gift-cards'
curl 'https://mcp.agenticterminal.ai/v1/merchants/bitrefill'
curl 'https://mcp.agenticterminal.ai/v1/categories'
curl 'https://mcp.agenticterminal.ai/v1/rails'
curl -X POST 'https://mcp.agenticterminal.ai/v1/merchants/bitrefill/verify?rail=lightning'
REST endpoints return the same JSON as the matching MCP tool (search_merchants ↔ GET /v1/merchants, get_merchant ↔ GET /v1/merchants/{id}, etc.), including the agent_identity field.
The MCP tool surface (paths A/B): search_merchants, get_merchant, verify_payment_endpoint, list_categories, list_rails, whoami.
2. Two orthogonal axes — always read both
Each merchant carries two independent signals. They do not imply each other.
OP trust tier (how verified the merchant is):
- Tier 1 — self-attested; merchant claims the rails, unverified.
- Tier 2 — operator-verified; the AT operator transacted with the merchant and confirmed at least one claimed rail (v1.x: any OP-credentialed counterparty can attest, not just AT).
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · 118 lines · 54 tokens per session scan A 4bf85736aaa5
agentic-terminal is a skill published in the GitHub repository observer-protocol/at-directory (0 stars, last pushed 14d ago), licensed MIT. It adds 54 tokens to every session and 1,800 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…