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/hirogakatageri/hirokata/shiftnpx skills add HirogaKatageri/hirokata --skill shiftgit clone --depth 1 https://github.com/HirogaKatageri/hirokataWrote 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/shift)<a href="https://agentmods.dev/skills/hirogakatageri/hirokata/shift"><img src="https://agentmods.dev/badge/skills/hirogakatageri/hirokata/shift.svg" alt="Measured on agentmods" 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 | $0.00126 | $0.07892 |
| Opus 5 | $0.00063 | $0.03946 |
| Sonnet 5 | $0.00025 | $0.01578 |
| Haiku 4.5 | $0.00013 | $0.00789 |
Grade A, and why
shift 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 — 635 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Guild Shift — working the board while you are away
Run until the next gate, then stop and notify.
That one sentence is the whole design. A requirement's work is cut into segments by its two gates, and a segment is by definition everything that can run without asking anyone anything. So the segment boundary and the stop boundary are the same line, and an unattended shift needs no separate notion of "how far may it go" — it goes to the next gate. Then it stops, records why, and waits.
Load guild:warehouse before the first turn. Every query on this page is raw SQL against
.guild/guild.db, and the six rules in that skill — hex transport, PRAGMA foreign_keys, never
parsing -m list positionally, reading views instead of re-deriving them, RETURNING on every
mutation, errors arriving on stdout — apply to every one of them. At 3am there is nobody to
notice a swallowed error.
This skill is guild:check-in with the human taken out of the middle. The dispatch protocol is
unchanged and is not restated here — how a batch is run, how an agent is spawned, and how a
result is recorded all live in guild:check-in, and you follow it exactly. What a shift adds is
a budget, a failure policy that never stops to ask, a commit per completed task, and a hard stop
at the gate.
Before you walk away — what you are authorizing
Say this out loud, before the first turn, so the user knows what they are leaving running. It is not a formality: they are handing an unsupervised process their working tree.
| A shift MAY | A shift MAY NOT |
|---|---|
| Claim any open, dependency-satisfied bounty | Approve a gate, or reject one |
| Run whole segments; compile and run workflows | Proceed past an unresolved gate |
| Retry a failed task once, with a fresh agent | Create a gate, or drop one |
Mark tasks failed or blocked, and move on |
Mark a requirement done past an unresolved gate |
| File bugs and review findings | Delete anything, or rewrite history |
Append to work_log and review_finding |
Push to a remote |
Commit per completed task, on guild/REQ-NNN |
Commit to the default branch |
| Change goals, projects or priorities | |
| Create a guild member |
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 First seen · 635 lines · 126 tokens per session scan A b81727d8aa25
shift is a skill published in the GitHub repository HirogaKatageri/hirokata (5 stars, last pushed 2d ago), licensed MIT. It adds 126 tokens to every session and 7,892 once invoked, about $0.0006 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 2026-09-04.
Other skills, from other repositories
unslop
Humanize LLM output so it reads like a careful human wrote it. Subtracts AI-isms (sycophancy, tricolons, em-dash overuse, "delve"/"tapestry"/"testament", hedging stacks, tidy five-paragraph shapes), engineers burstiness and calibrated uncertainty, and preserves technical accuracy. Supports intensity levels: subtle…
unslop-file
Humanize natural-language memory files (CLAUDE.md, todos, preferences, docs) by removing AI-isms and adding burstiness while preserving every code block, URL, path, command, and heading exactly. Two modes: --deterministic (fast, regex-based, no API) and LLM (default, calls Claude for rewrite). Humanized version…
unslop-commit
Rewrites commit messages so they sound like a careful human engineer wrote them. Strips AI/marketing slop ("comprehensive solution", "robust implementation", "leverage", "enhance", "seamlessly", "This commit..."). Keeps Conventional Commits format. Subject ≤72 chars (aim ≤50), imperative mood. Body only when "why"…
unslop-help
Quick-reference card for unslop modes, sub-skills, and slash commands. One-shot display, not a persistent mode. Trigger: /unslop-help, "unslop help", "what unslop commands", "how do I use unslop".
simulator-smart-forms-logic
Author and evolve the Corezoid backend that powers a Simulator.Company Smart Form. Use whenever the user wants a Smart Form to do anything beyond a static layout — load dynamic viewModel data, react to submits, drive page-level changes, trigger notifications, write back to actors — OR wants to modify existing backend…
simulator-finance
Simulator.Company financial management specialist. Use when the user wants to manage financial and metric accounts on actors (balances, counters, plan vs fact), record transactions, transfer value between accounts, work with currencies and account-name categories, track non-financial metrics, read balances and…