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 cyanheads/onebusaway-mcp-server --skill onebusaway-guidegit clone --depth 1 https://github.com/cyanheads/onebusaway-mcp-serverWrote 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/cyanheads/onebusaway-mcp-server/onebusaway-guide)<a href="https://agentmods.dev/skills/cyanheads/onebusaway-mcp-server/onebusaway-guide"><img src="https://agentmods.dev/badge/skills/cyanheads/onebusaway-mcp-server/onebusaway-guide/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/cyanheads/onebusaway-mcp-server/onebusaway-guide"><img src="https://agentmods.dev/badge/skills/cyanheads/onebusaway-mcp-server/onebusaway-guide.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 | $0.00057 | $0.01004 |
| Opus 5.5 | $0.00023 | $0.00402 |
| Sonnet 5.5 | $0.00011 | $0.00201 |
| Haiku 4.5 | $0.00006 | $0.00100 |
Grade A, and why
onebusaway-guide 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 3d 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.
How it starts
The opening of the file, as written. The whole thing — 49 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Scope
Use OneBusAway for service at known stops, route timetables, and available real-time telemetry. For an origin-to-destination journey, obtain a routing plan elsewhere before using these tools to check its stops and departures. Coverage belongs to the configured instance; a failed lookup does not establish that transit service is absent in a city.
Identifiers
| Identifier | Obtain from |
|---|---|
| Agency | onebusaway_list_agencies, or the agency attached to a discovered route |
| Stop | onebusaway_find_stops or onebusaway_search_stops; trip stop sequences also carry stop IDs |
| Route | Route discovery, agency route listings, a stop's served-route IDs, or an arrival |
| Trip | An arrival or schedule departure |
| Situation | Arrival situation IDs or the situations returned alongside arrivals; trip results also reference situations |
| Block | A non-null block ID from onebusaway_get_trip |
Keep the identifiers attached to the same instance and agency throughout the chain. A vehicle ID identifies a vehicle; use the trip ID returned with an arrival or vehicle record for trip lookup.
Gates
- Before choosing a boarding stop, reconcile its location, travel direction, served route, and arrival headsign with the rider's destination. Opposite directions can have similar stop names. When the available results cannot distinguish the intended stop, resolve that ambiguity before giving boarding advice.
- Before tracking a numbered route, identify its operating agency as well as its route ID. Use that agency for vehicle lookup; a display number alone does not establish identity.
- Before declaring a route or stop absent, try an appropriate alternate discovery path. Name-search support varies; an empty search is not coverage proof. Use coordinates near the service area, or discover the agency and browse its routes. Check
limitExceededbefore treating any returned list as complete. - Before calling an arrival late or early, check
predicted. Schedule-only arrivals do not establish a live delay, and a missing position does not establish that a bus is not running. Report the distinction to the rider. - Before interpreting an empty arrivals response as no service, check the requested window and then the dated stop schedule. An empty vehicle response also does not prove no service: the agency may lack real-time telemetry.
- Before concluding there are no disruptions, inspect notices about unresolved alert references. Missing alert detail is unknown, not an all-clear.
- Before presenting a local clock time, establish the agency timezone. Arrival and stop-schedule timestamps are epoch milliseconds; trip stop-sequence times are GTFS seconds from midnight. Keep extended service-day hours attached to their service day rather than silently wrapping them into the same day's early morning.
- Before using a block as a particular day's operating schedule, establish which service-calendar conditions apply. A block groups trips; its first returned configuration alone does not establish today's service.
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.
- 3d ago First seen · 49 lines · 57 tokens per session scan A bfb121729fbf
onebusaway-guide is a skill published in the GitHub repository cyanheads/onebusaway-mcp-server (1 stars, last pushed 4d ago), licensed Apache-2.0. It adds 57 tokens to every session and 1,004 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 2026-10-27.
Other skills, from other repositories
wikipedia-guide
Gates and identifier chains for the wikipedia tools. Covers switching language editions without reusing titles, disambiguation pages the article tools don't flag, section numbers versus sectionindex, coverage that differs by edition, and the checks to run before calling something absent or sourced. Use when a…
wikidata-guide
Use wikidata tools to identify entities, verify qualified claims, join external identifiers, and build SPARQL graph queries. Use for Wikidata research that needs identity disambiguation, statement evidence, or completeness checks.
openstates-guide
Identifier chains and gates for the openstates tools (US state legislatures via Open States). Use when tracing a state bill's fate, votes, or sponsors; finding who represents a location and their committees; finding hearings; or comparing states. Read before naming a bill, reporting who voted how, or concluding that a…
wsdot-guide
Plan Washington ferry trips with wsdot tools when a next-sailing, vehicle-eligibility, or capacity question needs schedule, service alerts, terminal space, and vessel evidence combined.
release-pr-review
Review pass on an open release PR (release/ → main) — the step between git-wrapup and release-and-publish when a project releases in gated release PR mode. Reads the PR's commit range through the code-simplifier lens plus a correctness review, verifies whatever an automated reviewer left on the PR, lands fixes as…
release-and-publish
Ship a release end-to-end across every registry the project targets (npm, MCP Registry, GitHub Releases for .mcpb bundles, GHCR). Runs the final verification gate, fast-forwards main when the release rode a release PR, creates the annotated tag on the commit main now points at, pushes commits and tags, then publishes…