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/davidmosiah/wellness-air/agents-mdgit clone --depth 1 https://github.com/davidmosiah/wellness-airWrote 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/davidmosiah/wellness-air/agents-md)<a href="https://agentmods.dev/instructions/davidmosiah/wellness-air/agents-md"><img src="https://agentmods.dev/badge/instructions/davidmosiah/wellness-air/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 5.1 | $0.00577 | $0.00577 |
| Opus 5 | $0.00289 | $0.00289 |
| Sonnet 5 | $0.00115 | $0.00115 |
| Haiku 4.5 | $0.00058 | $0.00058 |
Grade A, and why
wellness-air 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 5d 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 — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — wellness-air
Notes for AI agents and human contributors working on this repo.
What this is
A local-first MCP server that exposes air-quality readings (PM2.5, CO₂, AQI, etc.) to any MCP-aware agent. v0.1 ships full AirGradient support; AirThings/PurpleAir/IQAir/Awair are scaffolded.
Source layout
src/index.ts— MCP entry. Registers tools, runs stdio or Streamable HTTP.src/constants.ts— Versions, provider list, default ports/hosts.src/services/airgradient-client.ts— Public + owned AirGradient API client. Implements US EPA AQI conversion from PM2.5.src/services/{capabilities,privacy-audit,agent-manifest}.ts— Standard manifest surfaces.src/tools/air-tools.ts— All 10 tools registered with the MCP server.src/cli/commands.ts—doctor,setup,status,currentCLI commands.
Tool naming
All tools follow the air_<verb_or_noun> pattern. Resources use the wellness-air:// scheme.
Adding a provider in v0.2
- Create
src/services/<provider>-client.tsmodeled onairgradient-client.ts. Keep theAirReadingshape consistent. - Wire the provider into
air_current_readingandair_daily_summaryswitch insrc/tools/air-tools.ts. - Add provider env vars to
src/services/capabilities.tsconfigured-detection. - Update
src/services/agent-manifest.tsoptional_env. - Add provider notes to
src/services/privacy-audit.ts. - Bump
SERVER_VERSIONinsrc/constants.tsANDversionin bothpackage.jsonandserver.json.
Testing
npm testruns typecheck + build + smoke + metadata check.npm run smokeexercises the MCP tool surface against a public AirGradient sensor.npm run test:metadataensurespackage.json/server.jsonversions stay in sync.
Privacy contract
- AirGradient public reads = no auth, no PII, no token storage.
- Owned-sensor reads =
AIRGRADIENT_API_TOKENfrom env, never returned in tool output. - All cached data lives under
~/.wellness-air(override withWELLNESS_AIR_LOCAL_DIR). - No third-party telemetry. The only outbound call goes to the configured provider.
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.
- 5d ago First seen · 47 lines · 577 tokens per session scan A 4fce6b77b7bf
wellness-air AGENTS.md is an instructions file published in the GitHub repository davidmosiah/wellness-air (0 stars, last pushed 7d ago), licensed MIT. It adds 577 tokens to every session, about $0.0029 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 2026-08-31.
Other instructions, from other repositories
google-health-mcp AGENTS.md
Instructions for davidmosiah/google-health-mcp, covering agent development notes, scope, commands and rules.
wellness-nourish AGENTS.md
Instructions for davidmosiah/wellness-nourish, covering agent development notes, scope, commands and rules.
eight-sleep-mcp AGENTS.md
Instructions for davidmosiah/eight-sleep-mcp, covering agent development notes, scope, commands and rules.
wellness-cgm-mcp AGENTS.md
AGENTS.md instructions for davidmosiah/wellness-cgm-mcp, covering agents.md — wellness-cgm-mcp, what this is, source layout, mock mode and oauth flow (v0.1).
delx-living-body AGENTS.md
AGENTS.md instructions for davidmosiah/delx-living-body, covering agents.md — delx-living-body, what it is, tools (6), recommended agent flow and rules.
wellness-cycle-coach AGENTS.md
AGENTS.md instructions for davidmosiah/wellness-cycle-coach, covering agents.md — wellness-cycle-coach, what this is, source layout, phase model and adding new metrics.