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 gonnagetapower/kelvia-mcp --skill kelvia-boardgit clone --depth 1 https://github.com/gonnagetapower/kelvia-mcpWrote 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/gonnagetapower/kelvia-mcp/kelvia-board)<a href="https://agentmods.dev/skills/gonnagetapower/kelvia-mcp/kelvia-board"><img src="https://agentmods.dev/badge/skills/gonnagetapower/kelvia-mcp/kelvia-board/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/gonnagetapower/kelvia-mcp/kelvia-board"><img src="https://agentmods.dev/badge/skills/gonnagetapower/kelvia-mcp/kelvia-board.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 | $0.00069 | $0.00963 |
| Opus 5 | $0.00034 | $0.00481 |
| Sonnet 5 | $0.00014 | $0.00193 |
| Haiku 4.5 | $0.00007 | $0.00096 |
Grade A, and why
kelvia-board 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 12d 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 — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Working on a Kelvia board
Kelvia is a task board shared by a human and their agents. You have write access to it. Everything you do is attributed to your agent identity and shows up in the board's activity feed, where a human can review it and revert supported changes.
That attribution is the point of the product, and it changes how you should behave: the board is not scratch space. It is a record another person reads.
Before you write anything
Call list_boards and get_board first. A board has columns, and their names
are chosen by the user — do not assume "To Do / In Progress / Done". Read
list_board_columns and use the column IDs that exist.
If more than one board could be meant, ask which one rather than guessing. Filing work onto the wrong board is worse than asking.
Filing work
One task per unit of work a human would want to see separately. Resist both extremes: a single task called "implement the feature" is useless a week later, and fifteen tasks for one refactor is noise.
Write the title as the outcome, not the activity — "Search returns archived boards" rather than "Fix search". Put the reasoning, the reproduction, and the constraints in the description, where they survive.
When the user describes something ambiguous, file it and say what is unclear in the description rather than inventing a specification.
Keeping state honest
Move a task when its state actually changes, not at the end in a batch. If you start work, move it then. If you finish, move it then. A board that updates only at the end of a session is a board nobody trusts mid-session.
Use update_task to change the column, assignee, priority, or dates.
reorder_task positions within a column when order carries meaning.
Recording what happened
add_task_comment for reasoning: what you tried, what you rejected, what a human
should check. Comment when the outcome differed from the plan, not on every step.
add_task_worklog for time. Log the time actually spent on that task, once, when
the work is done. Do not invent durations; if you genuinely cannot tell, do not log.
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.
- 12d ago First seen · 96 lines · 69 tokens per session scan A 7ec18ea15aa5
kelvia-board is a skill published in the GitHub repository gonnagetapower/kelvia-mcp (1 stars, last pushed 28d ago), licensed MIT. It adds 69 tokens to every session and 963 once invoked, about $0.0003 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 skills, from other repositories
roadmap-planning-views
Creates and organizes focused planning views from a subset of roadmap tasks. Use when the user wants to build a task graph, analyze dependencies, identify ready or blocked work, or group near-term and future work into a focused planning workspace.
roadmap-task-flow
Captures, enriches, tags, and prioritizes roadmap tasks. Use when the user wants to note work quickly, add missing task details, organize a backlog, or reprioritize an existing list of tasks.
roadmap
Introduces roadmap-skill and routes roadmap work to the right workflow. Use when the user needs brief context about roadmap-skill, or wants help deciding between backlog task flow, planning views, and the web workspace.
roadmap-web-visualization
Opens and uses the local roadmap web workspace for visual review. Use when the user wants to inspect projects, tasks, kanban state, or planning views in the browser.
parallel-feature-development
Coordinate parallel feature development with file ownership strategies, conflict avoidance rules, and integration patterns for multi-agent implementation. Use this skill when decomposing a large feature into independent work streams, when two or more agents need to implement different layers of the same system…
bazel-build-optimization
Optimize Bazel builds for large-scale monorepos. Use when configuring Bazel, implementing remote execution, or optimizing build performance for enterprise codebases.