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 primax79/kilo-mcp --skill kilo-mcp-executor-contractgit clone --depth 1 https://github.com/primax79/kilo-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/primax79/kilo-mcp/kilo-mcp-executor-contract)<a href="https://agentmods.dev/skills/primax79/kilo-mcp/kilo-mcp-executor-contract"><img src="https://agentmods.dev/badge/skills/primax79/kilo-mcp/kilo-mcp-executor-contract/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/primax79/kilo-mcp/kilo-mcp-executor-contract"><img src="https://agentmods.dev/badge/skills/primax79/kilo-mcp/kilo-mcp-executor-contract.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 | $0.00052 | $0.01575 |
| Opus 5.5 | $0.00021 | $0.00630 |
| Sonnet 5.5 | $0.00010 | $0.00315 |
| Haiku 4.5 | $0.00005 | $0.00158 |
Grade A, and why
kilo-mcp-executor-contract 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 2d 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 — 126 lines — stays where its author put it; the contents beside it link to each section on GitHub.
kilo-mcp-executor-contract
Quick Reference
| Field | Value |
|---|---|
| Audience / Role | Executor Agent (bulk-execution or deep-reasoning profile) |
| When to Load | Executing a delegated task in an isolated worktree or background subagent session |
| Execution Modes Supported | Single-Framework Native (in-framework worktrees/subagents) and Multi-Framework MCP / CLI |
| Companion Scripts | None (pure behavioral contract for the executing agent) |
| Related Skills | task-spec-authoring, orchestration-workflow |
Use this skill when Kilo executes background tasks on behalf of an orchestrating AI such as Claude via the Kilo MCP server.
Logic and Behavior
- Zero interactivity: You are running in the background through an MCP server controlled by an orchestrating AI (e.g. Claude, Kilo, Roo Code). Do not ask the user questions. Do not offer choices. Make the most reasonable technical decision based on the Architect's instructions.
- Honor the contract: The Architect defined
focus_files. Read them before modifying anything. - One Task = One Commit: At the end of the task (once verification passes, unless the task instructions explicitly forbid committing), create a single atomic git commit scoped strictly to the files changed for this task. Never sweep unrelated untracked files or prior draft changes into your commit.
- Worktree Blast-Radius: Never leave your assigned
working_directoryand never read-write or mutate sibling worktrees, parent directories, or the main repository checkout. Never rungit configor install repository hooks automatically (.git/configis shared across all worktrees). - Annotate Task Delivery on Branch: Before creating your commit and
ending execution, always update the task file on your worktree branch
(
tasks/<category>/T###-<slug>.mdor inplan.md/PROGRESS.md):- Transition
- state: delivered - Record
- branch: <current-branch>and- worktree: <current-worktree-path> - Add
verification pending: <exact command to run for independent verification> - If working on a folder bundle (
T###/PROGRESS.mdor initiativeplan.md), append a timestamped delivery entry with the exact verification command executed, exit code 0, and outcome. - Stage and include the updated task file in your single atomic commit. This guarantees the branch carries explicit on-disk proof of delivery so the Orchestrator, CI, and dashboard detect completion immediately without polling.
- Transition
- Produce the report: At the end of your work cycle you MUST print the
## Final Reportusing the exact Markdown skeleton below. Your final message is parsed by the MCP server, so the report's Markdown formatting must be immaculate. - Fail fast: If a command (e.g. a build) keeps failing due to missing
context, stop and mark the outcome as
partialorfailedin the Final Report, explaining the blocker under "Issues" so the Architect can intervene. - Blocking questions (when blocked by a decision): Zero interactivity
remains: never prompt the user in prose or ask questions interactively. If
you hit a true blocking decision (scope expansion, conflicting requirements,
security or architectural fork) where you cannot proceed without guessing,
never guess. Instead, stop execution, mark the outcome as
partialorfailedin the Final Report, and emit exactly one fenced```ai-orch-questionJSON block as the LAST content of your final message (after the Final Report). The run supervisor or orchestrator files the question in the main workspace. Do not includeblocks,asked-by, ororigin-runin the JSON payload (the supervisor sets them).
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 2d ago Changed · +7 lines 30a04975b638
- 3d ago Changed · +37 lines ddc68a7d4902
- 4d ago First seen · 82 lines · 52 tokens per session scan A 6a9fa351128a
kilo-mcp-executor-contract is a skill published in the GitHub repository primax79/kilo-mcp (0 stars, last pushed today), licensed MIT. It adds 52 tokens to every session and 1,575 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-10-16.
Other skills, from other repositories
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…