cairn
Skill12 starsChanged 4 months agoMITView on GitHub
Quick natural-language logging and coaching for the Cairn training/nutrition tracker. Use when the user says things like "log my ride / run", "update my plan", "how am I tracking", "I'm down to X lb", "draft a meal plan", or "what's my workout today" over a configured Cairn MCP server.
5 other projects use a version of this.
Adopted by
Most recently taken up by
- hjxccc/cairntheir own version, first seen Oct 22, 2026
- CheswickDEV/cairntheir own version, first seen Oct 13, 2026
- CarriedWorldUniverse/nexustheir own version, first seen Oct 12, 2026
- z91772524-ai/pojia-next-mactheir own version, first seen Sep 28, 2026
- mhd-nasher/cairntheir own version, first seen Jun 17, 2026
Other versions5 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 5
- z91772524-ai/pojia-next-mac★ 39same name
- CheswickDEV/cairn★ 1same name
- hjxccc/cairn★ 1same name
- mhd-nasher/cairn★ 1same name
- CarriedWorldUniverse/nexus★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
The file
We split this skill into the rules it gives an agent
2 rules. 0 are common, 1 is distinctive, and 1 is very specific to this project.
0
Common
Most of the 4 edited versions we read keep these.
1
Distinctive
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 14
If the cairn tools (e.g. get_plan, log_activity) aren't available, tell the user to add it:
- Distinctive: line 99
After any write, read back what was stored so the user can confirm it landed.
“Common” is measured: kept by at least 6 in 10 of the 4 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Cairn quick actions
Cairn is the user's self-hosted training, nutrition & performance tracker. It exposes an MCP
server; you drive it with the cairn MCP tools. This skill maps everyday phrases to those tools
so the user can log and adjust things by talking instead of opening the web app.
Prerequisite: the cairn MCP server must be connected
If the cairn tools (e.g. get_plan, log_activity) aren't available, tell the user to add it:
Codex mcp add --transport http cairn http://localhost:8787/mcpReplace the URL with the host where Cairn runs: localhost, a LAN host, or an
HTTPS Tailscale/MagicDNS address. If that deployment has CAIRN_AUTH_TOKEN
enabled, configure the MCP client to send Authorization: Bearer <token> (or
use the token support provided by the client).
Intent → tool mapping
- "What's my workout today / what should I train?" → read
get_daily_sessionfirst; when nothing has been accepted, usepreview_daily_sessionfor the exact evidence-aware candidate Cairn would persist. This is read-only: callprepare_daily_sessiononly when the athlete explicitly starts or accepts it. "Show my weekly plan/template" →get_plan(orget_plan_dayfor one template day). - "Log my ride / run / swim / hike" (e.g. "did 2h on rolling trails, felt strong") →
log_activitywith the raw text intext. Cairn parses type/duration/distance/pace instantly, then (if enrichment is enabled) a background agent refines the entry and may distill a notable durable fact into memory. Read it back. - "What did I do on
<date>/ show a past session" →get_sessionwithdate: "YYYY-MM-DD"(returns that day's session + sets, or null). Useget_last_setfor an exercise's most recent set. - "Log a set" (e.g. "squat 195 for 8 at RIR 2") →
log_set(weight in lb; negative = assisted, null = bodyweight). Today's session is created automatically; the response flags a new est-1RM PR. - "How do I do X / form for X / cues for X" →
get_exercise(muscle group, injury constraint, form cues, plan appearances, est-1RM, recent sets). Lead with the constraint if there is one. - "Log my weight / I weighed in at X" →
log_weight(lb). It also updates the profile's current weight, so the goal check stays accurate. Uselist_weightfor the trend. - "How am I tracking / am I on pace" →
get_goal_check(TDEE + lean-safe feasibility) andget_progressfor the lifts they ask about (Epley est-1RM trend). Summarize honestly; if the goal is flagged aggressive, say so and quote the lean-safe recommendation. - "I'm down to 176 / push my goal date / change my weight / I live in Denver" →
set_profile(any subset). Store the usual base ashome_location; never request browser geolocation or treat it as a daily current-location picker. - "My priorities are … / endurance supports my main goal / I want to stay ready for 2-hour MTB
rides" →
set_training_intent. Keep the durable ordered priorities, endurance role, and sport-specific duration capability here. A temporary dated race belongs inset_endurance_goal; do not let the event silently replace the durable hierarchy. Read back withget_training_intent. - "Trail-system ride / trail MTB / downhill MTB / skiing in winter" → preserve the terrain or modality
in the
log_activitytext and in durable memory/intent when it describes identity. Trail/XC MTB includes climbs and descents and is not downhill-only; lift-served downhill, road, gravel, alpine skiing, Nordic skiing, touring, and unspecified skiing have different load meaning. Never invent a skiing subtype the athlete did not state. - "My run moved / what running still fits this week?" →
get_training_agenda. Treat run-plan days as suggested anchors: actual compatible runs close one weekly intention even on another day, completed work is not prescribed twice, and missed timing never creates catch-up volume. - "Update my plan / progress my targets for next week" →
draft_plan_updatewithagent: "auto"(uses the user's configured rotation), then route the result throughapply_proposal_with_autonomy. In Lead mode, bounded reversible changes land now or at the next natural boundary; structural changes announce first. Review mode still returns a draft that needs confirmation. - "Make next week 4/5/7 days / change my split" → restructure the plan. Use
set_planto replace the whole week (days not included are removed), orsave_plan_day/delete_plan_dayfor one day. Build sensible days that honor injuryconstraint_notes and carry weights over where it makes sense; show the proposed split and confirm before writing. - "Draft / refresh a meal plan for the week" →
draft_meal_plan(agent: "auto"). In Lead mode, the returnedautonomysidecar announces when it becomes current automatically; summarize the upcoming plan and date. In review posture it remains a draft and needs confirmation. - "Remember that I … / I prefer …" →
add_memory(dedupes exact repeats). Memory feeds every future coach prompt. "Fix / forget what you remember about …" →update_memory/delete_memory(the user can also curate this in the app's Me → Memory view). - "I'm travelling / trip to X on these dates / I tweaked my knee / rough week coming up" →
add_context_event(kind:trip|injury|life_event, withtitle,start_date, optionalend_date, and ametaobject — trip{location}, injury{area,severity}, event{impact}). The coach plans around these: deload/travel-friendly weeks over trips, de-load an injured area, ease volume during life events. A trip location is optional and overridesprofile.home_locationonly inside that trip's active date window; it never rewrites home.list_context_events({active}),update_context_event,delete_context_eventto review/curate. - Bloodwork / DEXA / a lab photo (vision) → analyze it yourself from the Codex client, then
add_health_record({kind, doc_date, summary, parsed}) to store the summary + markers so the coach factors them in (this MCP path records an analysis WITHOUT a binary; in-app uploads with auto-analysis happen in Me → Health).list_health_records/delete_health_recordto manage. - "Which agents are on / switch to round-robin / disable Grok" →
get_settings/set_settings(strategy: round_robin | random | priority;disabled_agents; weeklycoach_*schedule).
Rules
- Coaching outputs are ledgered proposals governed by Cairn's server autonomy policy. Under Lead mode, bounded reversible adaptations land at natural boundaries with rationale and Undo; structural changes announce first. Ask explicitly only for goal-identity, clinical, user-locked, high-risk, stale, or configured review-everything decisions. Never bypass the server policy with a direct apply.
- Respect injury constraints in the plan and in
list_memory; never suggest contradicting them. - For weight loss, defer to Cairn's lean-safe math (
get_goal_check) — don't endorse crash deficits. - After any write, read back what was stored so the user can confirm it landed.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

