Coros Training MCP Server
coros-training-mcp
PyPI
v0.3.1
Published by dholliday3 — no publish provenance, so origin is unverified, but the source is public: the repository link below is self-declared yet readable, so you can inspect the code before adopting it.
Model Context Protocol server for Coros with first-class running workout authoring, scheduling, and editing on top of training data, sleep, HRV, and activity exports.
The grade answers one question — how safe is this server for you to adopt — so it is computed in two auditable stages. Nothing below is an opinion or an LLM's guess; every line is a real term the deterministic engine applied, and the same input always yields the same number.
1. Threat score — 100 − 0 = 100. What the published surface and source actually contain:
The deterministic scan raised no scored threat in the surface it inspected — the threat score stayed at 100. Capability observations and advisory notes are recorded but never lower it.
2. Client adoption risk — 100 − 7 = 93. Three small, subtract-only factors that reflect your risk in adopting it — a clean scan proves less on a powerful, unverified or barely-inspectable package, so the grade says so plainly:
| Points | Adoption-risk factor |
|---|---|
| −6 | capability blast radius (high) — client exposure if the model is manipulated |
| −1 | publisher verification (public source) — no provenance, but the source is public and inspectable |
Capability observations and info notes are shown under Findings but never scored.
Open any row's finding below for the file, line and evidence behind a deduction.
Findings 4
high Shell/command execution in server code (auth/encrypted_store.py)MTC-SRC-002
In the server's implementation (`auth/encrypted_store.py:34`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: ss result = subprocess.run( ["ioreg", "-rd1", "-c", "IOPlatformExpertDevice"],
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server auth/encrypted_store.py
high Shell/command execution in server code (installer/assistants.py)MTC-SRC-002
In the server's implementation (`installer/assistants.py:191`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: if nothing's there. subprocess.run( ["claude", "mcp", "remove", MCP_ENTRY_NAME], capture_output=True
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server installer/assistants.py
high Shell/command execution in server code (installer/smoke.py)MTC-SRC-002
In the server's implementation (`installer/smoke.py:53`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: try: proc = subprocess.Popen( server_command, stdin=subprocess.PIPE, std
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server installer/smoke.py
high Shell/command execution in server code (traininghub_live_builder_catalog.py)MTC-SRC-002
In the server's implementation (`traininghub_live_builder_catalog.py:47`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: str) -> str: return subprocess.check_output( [ "security", "find-generic-password",
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server traininghub_live_builder_catalog.py
Tools 29
Each tool and what it can reach — statically extracted from the published source.
get_activity_detailingests untrusted inputget_daily_metricsingests untrusted inputget_sleep_dataingests untrusted inputget_workoutingests untrusted inputauthenticate_corosno sensitive capabilityauthenticate_coros_mobileno sensitive capabilitycheck_coros_authno sensitive capabilitycreate_bike_workoutno sensitive capabilitycreate_run_workoutno sensitive capabilitycreate_strength_workoutno sensitive capability
Show 19 more tools ↓
create_workoutno sensitive capabilitydelete_workoutno sensitive capabilityexport_activity_fileno sensitive capabilityget_bike_workout_schemano sensitive capabilityget_run_workout_schemano sensitive capabilityget_strength_workout_schemano sensitive capabilityget_workout_builder_catalogno sensitive capabilitylist_activitiesno sensitive capabilitylist_exercisesno sensitive capabilitylist_planned_activitiesno sensitive capabilitylist_scheduled_workoutsno sensitive capabilitylist_workoutsno sensitive capabilitymove_scheduled_workoutno sensitive capabilityremove_scheduled_workoutno sensitive capabilityreplace_scheduled_workoutno sensitive capabilityschedule_workoutno sensitive capabilityupdate_bike_workoutno sensitive capabilityupdate_run_workoutno sensitive capabilityupdate_workoutno sensitive capability
What this scan could not see
- Tools were statically extracted from the published source (29 recovered), not enumerated from a running server. Tool-poisoning, Unicode-smuggling, capability and toxic-flow analysis ran on this inferred surface, but a mis-parsed registration could be missed or mis-attributed, so tool-derived findings are capped below “confirmed”. To grade the real runtime surface, scan the running server: --command "npx -y <package>".
Versions 3
Scan history per published version. The engine is deterministic — the same version always yields the same score, so a changed score means the package itself changed.
| Version | Score | Findings | Engine | Scanned |
|---|---|---|---|---|
v0.3.1 latest |
A 93/100 | 4 | 1.13.0 | 2026-08-25 |
v0.3.0 |
A 93/100 | 4 | 1.12.1 | 2026-08-07 |
v0.2.0 |
A 93/100 | 4 | 1.12.1 | 2026-07-27 |
Embed this score
Show this server's live Trust Score in your README, docs or website. The badge is served straight from the registry and updates automatically after every rescan — no API key needed. It links back to this page, so anyone who sees the grade can also read the findings behind it instead of taking a number on faith.
Verify this score yourself
The score above is reproducible: the same package version always yields the same result. Run it locally or over the free API — no account, no LLM, fully deterministic.
npx mcptrustchecker scan coros-training-mcp --online --registry pypiMore in Lifestyle & Local
MCP server for Australian address search and validation via Addressr
Admissions data & match scores for law, medical, dental, MBA, pharmacy, vet, optometry schools
Live AI data for agents: model releases, regulations (EU AI Act), GenAI glossary, daily news.
AirTreks MCP Server — complex flight routing intelligence for AI agents. Plan multi-city trips, validate alliance routings, get carrier recommendations, and create trip ideas.
Physician-reviewed medical opinions and prescriptions for AI agents.
Thin MCP server + CLI that proxy the atrium /api/v1 observability surface using an oc_live_ Bearer key. Talks HTTP only — no Convex imports.