kilter-mcp
This is a read-only MCP server that lets AI agents explore a Kilter Board climber's own logbook data — sessions, sends, projects, grades and trends — without writing anything to the account.
kilter_get_logs — fetch raw logbook entries, newest first, with filters for angle, topped status, limit, and date range.
kilter_get_sends — list topped climbs (sends/flashes), newest first, with angle and date filters.
kilter_get_projects — see unfinished climbs (attempted but never topped at a given angle), sorted by recency and attempts.
kilter_get_summary — get a compact overview: totals, angles, date range, sessions, hardest send/flash, sends per grade.
kilter_get_grade_pyramid — view send counts per grade with flash rates, optionally filtered by angle.
kilter_get_hardest_sends — show hardest unique sends / personal bests.
kilter_get_sessions — see climbs grouped by calendar day (like “How did my last session go?”).
kilter_get_progression — track month-by-month or week-by-week improvement.
kilter_get_angle_stats — compare performance across wall angles (e.g. 20° vs 40°).
All tools are read-only, work with your own Kilter credentials, group data by your timezone, and are usable from MCP clients like Kiro, Claude, Codex, Cursor, and more.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@kilter-mcp What are my current projects at 30°?"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
kilter-mcp
An unofficial MCP server that lets your AI agent read and analyse your own Kilter Board logbook.
Add it to Kiro (or another MCP client), give it your Kilter login, and then just talk to the agent:
How did my last Kilter session go? What are my current projects at 30°? Compare my climbing at 20° and 40°. Show my grade pyramid. What are my hardest sends? Am I progressing?
Read-only. Nothing is written to your Kilter account.
Add to Kiro + Other Clients (Claude, Codex, Cursor etc.)
Click the button (Kiro shows a confirmation dialog first), then jump to step 3. Or do it by hand:
Open your MCP config in Kiro: command palette → Kiro: Open user MCP config (JSON) (or the workspace one,
.kiro/settings/mcp.json).Add the server:
{
"mcpServers": {
"kilter": {
"command": "uvx",
"args": ["kilter-mcp"],
"env": {
"KILTER_USERNAME": "${KILTER_USERNAME}",
"KILTER_PASSWORD": "${KILTER_PASSWORD}"
},
"disabled": false,
"autoApprove": []
}
}
}Provide your Kilter credentials. Either:
set
KILTER_USERNAMEandKILTER_PASSWORDas environment variables (Kiro expands${VAR}after you approve the variables in Settings → Mcp Approved Env Vars; see how to set them on macOS, Linux and Windows), orreplace the
${...}placeholders with the literal values. Only do this in the user config (~/.kiro/settings/mcp.json), never in a workspace config you might commit.
Save. Kiro starts the server and the
kilter_*tools appear. Ask away.
Something not working? With the same variables exported in a terminal, run
uvx kilter-mcp --check. It logs in, counts your logbook and prints a short report with no
secrets or entries in it, which you can paste into an issue.
Requirements: uv installed (uvx comes with it). Python is
handled by uv.
Kiro CLI users can do the same with
kiro-cli mcp add --name kilter --scope global --command uvx --args kilter-mcp --env KILTER_USERNAME=... --env KILTER_PASSWORD=....
Optional: timezone
Sessions are grouped by calendar day in your timezone, which is detected from the machine
running the server. If that is wrong (for example a remote or containerised host), add
"KILTER_TIMEZONE": "Europe/Rome" (any IANA name) to the env block. Every response states
the timezone it used.
Other clients
kilter-mcp is a standard local MCP server, so it also works in Claude Desktop, Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI and Zed. Copy-paste configs for each are in docs/clients.md. (ChatGPT's chat interface only accepts remote HTTP servers, so it is not supported directly; details in the same document.)
Related MCP server: strava-mcp
Example questions
Show my latest Kilter session.
What did I climb last Saturday?
List my sends from this month at 40°.
What are my current projects?
Which projects have I put the most attempts into?
Show my grade pyramid at 30°.
What is my flash rate per grade?
What are my five hardest sends?
Compare my climbing at 20° and 30°.
How has my climbing progressed over the last six months?Tools the agent gets
Tool | What it returns |
| Logbook entries (filter by limit, angle, topped, date range) |
| Topped climbs, newest first |
| Climbs attempted but never topped at that angle |
| Totals, angles, date range, hardest send/flash, sends by grade |
| Sends per grade with flash rates |
| Hardest unique sends |
| Entries grouped by day |
| Month-by-month or week-by-week trend |
| Per-angle comparison |
All tools are read-only and only ever see the account whose credentials you configured.
How the data is interpreted
One logbook entry is one climb at one wall angle on one date, with the number of tries in that entry. Re-logging a climb creates a new entry.
statusisflash(topped first try),send(topped) orattempt(not topped).A project is a
(climb, angle)pair with attempts but no top at that angle.Grades are Kilter's current consensus grade for the climb at that angle.
gradeis the V-scale andfont_gradethe Font scale; the rawdifficulty_id(1-39) is always included. The grade table is fetched from Kilter at startup with an embedded fallback.If you rated a climb yourself in the app,
my_gradeandmy_rating(1-5 stars) carry your opinion; otherwise they are null.Dates are reported in your timezone (see above). A session is one calendar day.
Security and privacy
Your username and password are only used to obtain an OAuth token from Kilter's login server (
idp.kiltergrips.com). They never leave your machine otherwise.Tokens are kept in memory for the life of the MCP process and are never written to disk.
Credentials and tokens are never included in tool output, logs or error messages.
The server makes requests only to
idp.kiltergrips.comandportal.kiltergrips.com.Also available in the MCP Registry as
io.github.wmmg101/kilter-mcp.Prefer the
${KILTER_PASSWORD}form so the password lives in your shell environment or secret manager rather than in a JSON file.Your logbook is fetched at most once a minute, however many tools the agent calls, and Kilter's rate-limit responses are respected.
Full details and how to report a problem privately: SECURITY.md.
Disclaimer
kilter-mcp is an unofficial community project and is not affiliated with or endorsed by Kilter Grips. It talks to the same endpoints the Kilter app uses; these are not a documented public API and may change or stop working at any time. Kilter's Terms of Use restrict access to their services outside the official apps; use this project at your own discretion, only with your own account, and keep request volume low (the server fetches your logbook once per tool call).
Contributing
Issues and pull requests are welcome. See CONTRIBUTING.md for the
development setup (git clone, uv sync, uv run pytest), architecture notes, and how to
add a tool.
License
Available Tools
9 toolskilter_get_angle_statsARead-only
Return per-wall-angle statistics so angles can be compared.
Use when the user asks to compare angles (e.g. 20° vs 40°), which angle they climb hardest or most at, or flash rate by angle. One row per angle with sends, flashes, attempts, sessions, hardest send/flash and sends per grade.
Inputs: no named parameters. Returns: object
kilter_get_grade_pyramidARead-only
Return the user's send pyramid: sends per grade, hardest first, with flash rates.
Use when the user asks about their grade pyramid, distribution of grades, flash rate per grade, or how solid they are at a level. Optional wall-angle filter.
Inputs: angle. Returns: object
kilter_get_hardest_sendsARead-only
Return the user's hardest sent climbs (unique per climb and angle), hardest first.
Use when the user asks about their hardest sends, best climbs, max grade, or personal bests. Optional wall-angle filter; limit defaults to 10.
Inputs: angle, limit. Returns: object
kilter_get_logsARead-only
Return the user's Kilter logbook entries, newest first.
Use when the user asks about recent Kilter climbs, attempts, what they climbed on a date, or wants raw history. Each entry is one climb at one wall angle on one date with the number of tries in that entry. Filters: limit (default 25, max 500), angle (degrees), topped (true=only sends, false=only unsuccessful attempts), start_date/end_date (YYYY-MM-DD, inclusive).
Inputs: angle, limit, topped, end_date, +1 more. Returns: object
kilter_get_progressionARead-only
Return climbing progression over time, oldest period first.
Use when the user asks how they have progressed, trends, whether they are improving, or for a month-by-month/week-by-week view. period is 'month' (default) or 'week'. Each period has sessions, sends, unique climbs, flashes, attempts and hardest send/flash. Optional wall-angle filter.
Inputs: angle, period. Returns: object
kilter_get_projectsARead-only
Return the user's projects: climbs attempted but never topped at that wall angle.
Use when the user asks what they are working on, unfinished climbs, or what to try next. A climb is a project per (climb, angle); sending it at another angle does not remove it. Sorted by most recently tried, then most attempts. Optional angle filter.
Inputs: angle, limit. Returns: object
kilter_get_sendsARead-only
Return climbs the user has topped (sent), newest first.
Use when the user asks about their sends, ticks, completed climbs or flashes. Entries with status 'flash' were topped on the first try. Filters: limit (default 25, max 500), angle (degrees), start_date/end_date (YYYY-MM-DD).
Inputs: angle, limit, end_date, start_date. Returns: object
kilter_get_sessionsARead-only
Return recent climbing sessions (entries grouped by calendar day), newest first.
Use when the user asks how their last session went, what they climbed yesterday, or wants to compare sessions. Each session lists angles, sends, flashes, attempts, hardest send and every climb logged that day. limit defaults to 5.
Inputs: limit. Returns: object
kilter_get_summaryARead-only
Return a compact overview of the user's whole Kilter history.
Use first when the user asks a broad question ("how is my climbing going?", "summarise my Kilter account") or when you need totals: entries, sends, flashes, attempts, angles climbed, date range, session count, hardest send/flash and sends per grade.
Inputs: no named parameters. Returns: object
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 9 tools
Each tool targets a distinct analytical view (sends vs logs vs projects vs sessions), but sends, logs, and hardest_sends overlap in the 'climbed successfully' space. Descriptions clarify the differences well, so confusion is unlikely but possible for an agent searching for a send list.
All tools follow the exact same kilter_get_<noun> pattern, making the naming scheme highly predictable and easy to navigate. No mixed conventions or verb variations exist.
Nine tools is well within the ideal range for a focused analytics server. Each tool covers a distinct query mode without unnecessary bloat or obvious redundancy.
The tool surface covers the key analytical queries for a Kilter climbing history: overall summary, per-angle stats, progression, sessions, projects, sends, and hardest sends. Minor gaps exist such as no per-climb detail view or direct grade-based filtering of send lists, but these are workable via existing tools.
Maintenance
Related MCP Connectors
Your own cycling data for your AI assistant: rides, power, climbs and routes. Read-only.
Read-only AI access to Osboon business card analytics, viewers, links, connections and contacts.
Read-only access to your PumpX workout log: sessions, lift history, PRs, routines, measurements.
Ask your AI about your Airbnb and VRBO listings, bookings, guest messages and reviews. Read-only.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceGives LLM agents read-only access to your WHOOP data — recovery, sleep, strain, workouts, and profile.14 npmMIT
- AlicenseAqualityBmaintenanceEnables AI assistants to read your Strava training data and answer plain-language questions about activities, splits, and totals across sports like running, skiing, and hiking.48 npmMIT
- AlicenseAqualityBmaintenanceEnables AI agents to read and analyze your Boardsesh climbing logbook across boards like Kilter, Tension, and MoonBoard, with read-only access and optional credentials for personalized recommendations.1225 PyPIMIT
- AlicenseAqualityBmaintenanceEnables AI assistants to read and analyze athlete training data, including fitness summaries, activities, and profiles, to help plan training.369 npmMIT