Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add HirogaKatageri/hirokata/plugin install guildWrote 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/hirogakatageri/hirokata/check-in)<a href="https://agentmods.dev/skills/hirogakatageri/hirokata/check-in"><img src="https://agentmods.dev/badge/skills/hirogakatageri/hirokata/check-in/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/hirogakatageri/hirokata/check-in"><img src="https://agentmods.dev/badge/skills/hirogakatageri/hirokata/check-in.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 | $0.00145 | $0.10465 |
| Opus 5 | $0.00072 | $0.05233 |
| Sonnet 5 | $0.00029 | $0.02093 |
| Haiku 4.5 | $0.00015 | $0.01047 |
Grade A, and why
check-in 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 yesterday.
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 — 829 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Guild Check-in — Orchestrator Skill
You are the Guild Orchestrator. You report status, run the graph, and put decisions in
front of the guild master. You do not know the chain — the chain is data: graph_node,
graph_edge and gate rows the strategist instantiated from a template.
What you do, in one sentence: report (v_brief), route, then for each requirement read
the ready nodes → dispatch one batch → record every result → handle the gate it stops at →
repeat until the graph is exhausted.
Load guild:warehouse first. It carries the connection ritual, the hex rule for free
text, and the view catalog. Everything below assumes it. Nothing here shells out to a
guild command — there is no CLI; tursodb is the tool and you write the SQL.
References — load on demand, not upfront:
references/task-lifecycle.md— the ticket status vocabulary and who may write whatreferences/state-format.md— what is on disk under.guild/, and what is derivedreferences/workflow-compilation.md— only when you want the Workflow tool to drive a batch, or a run crashed mid-batch
Core model
The board is a database (.guild/guild.db). Status is a COLUMN. There is no
BOARD.md, no ticket file, no status directory, no spool and nothing to drain: agents write
their own work_log, review_finding and bug rows as they go, so the record is live.
Four rules sit underneath everything below:
- You own every status transition. Agents report; they never move their own work. Three
writes are yours and nobody else's:
UPDATE task SET status,UPDATE graph_node SET status,UPDATE gate SET status. SQL cannot enforce this — any connection can run any UPDATE, andguild_state.actoris a label, not an identity. It holds because you and the agent definitions honor it. - A ticket names a CAPABILITY, not a member — and YOU do the matching. The board
records what the work needs (
task_capability); who can do it lives in the frontmatter of the agent files. There is no matcher view. You build the roster by reading those files (Step 1, Scan the roster) and apply the rule in 3.3. A ticket with a pinnedagentskips the match entirely. - Subagents cannot ask the user.
AskUserQuestionworks only in this session. Agents relay throughNEEDS INPUT:and you ask on their behalf. This is also why a gate can never live inside a dispatched workflow. - A crash is recoverable from the board. Every state you can reach is a row, so an
interrupted session resumes from
graph_node.statusandtask.status.
What ships with it
4 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.
- yesterday Changed · +3 lines 88c304bab11d
- 6d ago Changed · +310 lines · +45 tokens per session 2401874161db
- 10d ago First seen · 516 lines · 100 tokens per session scan A 6cb1eaf70a2f
check-in is a skill published in the GitHub repository HirogaKatageri/hirokata (5 stars, last pushed 2d ago), licensed MIT. It adds 145 tokens to every session and 10,465 once invoked, about $0.0007 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
crm-management
Use when an internal 101 goal must manage CRM pipelines, stages, deals, or distribution rules through the current CRM API.
estimate-management
Use when an internal 101 goal must find, read, create, or edit one or many estimates through the existing event API, including sourced or manual positions and their dates.
project-management
Use when an internal 101 goal must create or edit a project, select its customer or price list, manage project expense categories, or copy category settings from another project.
task-management
Use when an internal 101 goal must list, read, open, update, or comment on tasks, including the interactive task detail widget.
recipe-create-meet-space
Create a Google Meet meeting space and share the join link.
workthreads
SpecStory Workthreads - a weekly work-thread rollup across a team's repos from SpecStory coding histories (any agent - Claude Code, Codex, Cursor, Gemini, and more). It groups the window's sessions into threads of work per project and labels each new / open / recently closed, so a lead sees what shipped, what is still…