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 agentmods add skills/klimentij/klimkit/klimkit-implementnpx skills add klimentij/klimkit --skill klimkit-implementgit clone --depth 1 https://github.com/klimentij/klimkitWhat 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 | $0.00074 | $0.00975 |
| Opus 5 | $0.00037 | $0.00487 |
| Sonnet 5 | $0.00015 | $0.00195 |
| Haiku 4.5 | $0.00007 | $0.00097 |
Grade A, and why
klimkit-implement 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 3d 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 — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Klimkit Implement
Use this as the skills-first replacement for the old global AGENTS.md implementation workflow. Keep the work in the main agent by default. Do not use subagents for exploration, checklists, debugging, security, or reflection unless the user explicitly asks; the only built-in subagent use is the final-review gate.
Defaults
- Treat Klimkit as a Vercel Skills CLI package first: install and update the root
skills/library withnpx skills. - Use normal Codex app or CLI sessions plus installed Klimkit skills for day-to-day work.
- Keep task evidence in the project repository, not in generated home files.
- Treat older Klimkit runtime machinery as deprecated legacy material. Do not route new workflows through it unless the user explicitly asks to maintain legacy code.
Workflow
- Intake
- Read the latest user request, repo instructions (
AGENTS.md), relevantdocs/work/LOGs (descend selectively), docs, and nearby code. - If the repo lacks the
docs/work/layout, useklimkit-setup. - Decide whether the work is implementation, debugging, release, review, research, UI/proof, or security-sensitive.
- Read the latest user request, repo instructions (
- Explore
- Use
klimkit-code-explorerinline when the code path is unfamiliar, broad, or risky. - Use direct repo reads for simple tasks.
- Use
- Checklist
- Use
klimkit-checklisterbefore code changes for implementation work. - Treat the checklist as blocking unless the user changes scope.
- Use
- Implement
- Choose the narrow method skill when helpful:
klimkit-diagnose,klimkit-tdd,klimkit-walkthrough,klimkit-report-server, orklimkit-create-worktree. - Keep edits surgical and aligned with existing repo style.
- Do not route new work through deprecated runtime, Switchboard, sync scripts, or plugin machinery unless explicitly maintaining legacy code.
- If a task asks to migrate old Klimkit runtime behavior into skills, make the target skill own any reference scripts, templates, and setup instructions it needs.
- Choose the narrow method skill when helpful:
- Verify
- Run checks that match the checklist and blast radius.
- Record exact commands, important outputs, skipped checks, and remaining risk.
- For UI/proof work, produce inspectable report evidence when the repo workflow requires it.
- Security
- Use
klimkit-security-auditorbefore completion when the change touches auth, secrets, permissions, containers, network exposure, infrastructure, user data, or unsafe defaults.
- Use
- Reflect
- Use
klimkit-reflectorafter verification and before final review for non-trivial work. - If reflection exposes a material gap, fix it and rerun impacted checks.
- Use
- Final Review
- Draft the exact final response.
- Run fresh final-review passes with
klimkit-final-reviewer. - Prefer 3 independent final-review subagents when subagents are available and the user has not forbidden them. Give each only the original request, checklist, changed files, verification evidence, reflection note, proof artifacts, and exact draft response.
- If subagents are unavailable or forbidden, run
klimkit-final-reviewerinline and state that limitation. - Do not claim completion until final review returns
READY FOR USERor all remaining gaps are explicitly reported.
- Report
- Tell the user what changed, what passed, what was skipped, and how to inspect the result.
- For commits on
mainin this repo, create or update the latest GitHub release according to repo instructions.
What ships with it
2 files 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.
- 3d ago First seen · 62 lines · 74 tokens per session scan A e45effe009c0
klimkit-implement is a skill published in the GitHub repository klimentij/klimkit (7 stars, last pushed 23d ago), licensed MIT. It adds 74 tokens to every session and 975 once invoked, about $0.0004 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
happiness-skill
当用户问「怎么才能更幸福/为什么得到了还不满足/怎么减少焦虑」时调用。 核心理念: 幸福是缺憾感清空的默认状态, 是可训练的技能; 欲望是与自己的契约(得到前不快乐), 同时只留一个重大欲望; 活在当下。 不适用于: 临床抑郁等需要专业治疗的场景(本书方法不能替代医疗)。 Triggers: 幸福/不快乐/欲望/焦虑/知足/活在当下/happiness/desire/anxiety.
agr-cli
Install, share, sync, and create AI agent skills across coding tools (Claude Code, Cursor, Codex, OpenCode, Copilot, Pi) using the agr CLI. Use whenever the user mentions agr, agr.toml, agr.lock, agrx, or asks to: add a skill ("install the pdf skill", "agr add ..."), sync agent resources across tools, share skills…
agr-release
Release process for the agr package. Handles version bumping (major/minor/patch/beta), changelog updates, pre-release quality checks, git tagging, and monitoring the GitHub Actions publish pipeline. Use this skill whenever the user wants to cut a release, bump the version, publish to PyPI, or asks about the release…
hook-craft
Specializes in chapter openings (hooks) and chapter endings (pulls). Every chapter must start with a reason to keep reading and end with a reason to turn the page. Runs after prose-craft, before chaos-engine. The skill that prevents the reader from putting the book down.
kafka-topic-audit
Audit all Kafka topic configurations against production best practices using the Lenses MCP server. Checks replication factor, retention, partitions, compaction, naming conventions, orphaned topics and missing metadata. Use when user says "audit my topics", "check topic configs", "topic health check" or asks about…
book-genesis
Use when the user wants to create, plan, draft, evaluate, revise, package, or run a complete AI-assisted book pipeline. Book Genesis turns a one-line idea into a book project with intake, foundation, architecture, drafting, adversarial audit, Genesis Score, and editorial package artifacts.