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 kilroy-sh/kilroy --skill using-kilroygit clone --depth 1 https://github.com/kilroy-sh/kilroyWrote 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/kilroy-sh/kilroy/using-kilroy)<a href="https://agentmods.dev/skills/kilroy-sh/kilroy/using-kilroy"><img src="https://agentmods.dev/badge/skills/kilroy-sh/kilroy/using-kilroy/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/kilroy-sh/kilroy/using-kilroy"><img src="https://agentmods.dev/badge/skills/kilroy-sh/kilroy/using-kilroy.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 | $0.01410 |
| Opus 5.5 | $0.00021 | $0.00564 |
| Sonnet 5.5 | $0.00010 | $0.00282 |
| Haiku 4.5 | $0.00005 | $0.00141 |
Grade A, and why
using-kilroy 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 2mo 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 — 129 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Kilroy is your team's shared memory — posts from teammates, their coding agents, and past sessions. It persists across people, sessions, and tools. It is NOT your private memory: what you post, the whole team sees.
If Kilroy tools fail or return auth errors, ask the user to reconnect from the Kilroy web dashboard (or re-add the connector in Settings).
Project routing
Every Kilroy tool call requires an explicit project parameter
(account/slug). There is no default — never guess.
- Call
kilroy_list_projectsto see what the user can access. - One project → use it. Several → ask the user once which project this conversation is about.
- Pass that value as
projecton every call for the rest of the conversation.
Reading — answer from the team's memory
When the user asks what the team knows, decided, or learned — or starts work someone may have touched before — check Kilroy before answering from general knowledge.
kilroy_searchwith a few focused terms; omitqueryto browse recent posts.kilroy_tagsshows what topics exist.- Read the top posts fully with
kilroy_read_post— comments often carry corrections and updates. - Cite post URLs in your answer so the user can verify and read on.
- Prefer recent posts;
created_attells you how stale a claim may be. - Posts conflict? Prefer the most recent, tell the user, and comment on the stale post noting the discrepancy.
- A post records what was true when written. For claims about live systems or code, say what the post claims and when it was written — don't present it as current fact you verified.
Writing — capture what the team will need
Would a teammate or future session benefit from knowing this? If yes,
post it — don't ask permission, just mention the post (with its URL)
afterward. Post at the first meaningful insight; refine later with
kilroy_edit_post (targeted find/replace) or kilroy_update_post
(full rewrite). Capture when:
- A decision lands — record the why and the alternatives dismissed,
not just the outcome. Proposals use the same shape, tagged
proposal, updated when the decision lands. - Research or analysis concludes — findings are expensive to reproduce.
- A meeting or discussion produces outcomes — decisions made, owners assigned, constraints surfaced. Capture the durable part, not minutes.
- A procedure gets figured out — steps someone will repeat.
- The user shares reusable context — vendor limits, org constraints, domain facts.
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.
- 2mo ago First seen · 129 lines · 52 tokens per session scan A ba8b75d4e9f3
using-kilroy is a skill published in the GitHub repository kilroy-sh/kilroy (10 stars, last pushed 1mo ago), licensed MIT. It adds 52 tokens to every session and 1,410 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 2026-08-31.
using-kilroy compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| using-kilroykilroy-sh/kilroy | Skill | 10 | 1mo ago | MIT | A | 52 | Claude Code |
| knowledge-ingestopenclaw/openclaw | Skill | 394k | yesterday | MIT | A | 80 | Claude Code, Codex |
| mem0-oss-to-platformmem0ai/mem0 | Skill | 68k | 2d ago | Apache-2.0 | A | 217 | Claude Code, Codex |
| forgetasgeirtj/system_prompts_leaks | Skill | 71k | 2d ago | CC0-1.0 | A | 74 | Claude Code, Codex |
| Cortexdanielmiessler/LifeOS | Skill | 20k | yesterday | MIT | A | 184 | Claude Code |
| memoryopensquilla/opensquilla | Skill | 7.1k | 6d ago | Apache-2.0 | A | 44 | Claude Code, Codex |
| ha-data-storesshiwenwen/hope-agent | Skill | 1.9k | yesterday | MIT | A | 115 | Claude Code, Codex |