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 SoliEstre/EstreGenesis --skill routinizegit clone --depth 1 https://github.com/SoliEstre/EstreGenesisWrote 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/soliestre/estregenesis/routinize)<a href="https://agentmods.dev/skills/soliestre/estregenesis/routinize"><img src="https://agentmods.dev/badge/skills/soliestre/estregenesis/routinize/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/soliestre/estregenesis/routinize"><img src="https://agentmods.dev/badge/skills/soliestre/estregenesis/routinize.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 | $0.00148 | $0.01027 |
| Opus 5 | $0.00074 | $0.00513 |
| Sonnet 5 | $0.00030 | $0.00205 |
| Haiku 4.5 | $0.00015 | $0.00103 |
Grade A, and why
routinize 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 10d 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 — 32 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/routinize — codify a recurring practice (관행화)
Turn "work I keep doing the same way" into a durable Greatpractice entry — without over-codifying. This is the model-invoked front door to Greatpractice's capture → mature → codify pipeline (Greatpractice.md §1.2). The full normative model — tiers, maturity score, notability gate, phronesis boundary — lives in Greatpractice.md; this skill is the actionable procedure, not a restatement.
Core discipline — capture is immediate, promotion is deferred (§5.4). A single session's repetition is usually the same trigger coordinate (§5.2), so most invocations end at capture/accumulate; only genuinely independent recurrence produces a draft. The skill never registers an enforcing hook on its own authority — that is a user decision.
Procedure
-
Scan for recurrence — over the recent context, find work patterns that repeated: the same fix applied ≥2×, the same multi-step sequence run again, a "when X, also do Y" coupling, a mistake made → corrected → nearly repeated. Name each candidate pattern in one line.
-
Phronesis gate — what NOT to codify (§1.4 / §5.3). Drop any candidate that is judgement-heavy (outcome doesn't reduce to one metric), rare-context (long-tail work phase), or high-context-dependent (the same surface behavior means different things by context). These stay in the agent's instinct — codifying them causes phronesis atrophy. Keep a borderline-but-worth-capturing candidate, but mark
phronesis_boundary: true(advisory only, never blocking). -
Route each surviving candidate (§5.4):
- New pattern → write
memory/feedback_<slug>.md: minimal frontmatter (trigger, timestamp,trigger_source),lifecycle: probation, blameless voice (state the pattern, not blame). Eager capture — cost 0, no approval needed. - Already captured → append this occurrence to the existing feedback's illustration section + record its
(work-domain × phase × time-of-day)coordinate; bump the significant-coverage counter (substantive appends only). - Notability gate met — only if the pattern now has ≥3 substantive occurrences across ≥2 distinct coordinates (3× at the same coordinate = 1 trigger, does not qualify) AND a verifiable effect is definable (§5.2) → draft
greatpractice/_propose/<slug>.draft.mdwith the full §3 frontmatter, a tier guess (§2: macro governance / mezzo procedure / micro atom — most are mezzo), and the phronesis verdict.
- New pattern → write
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.
- 10d ago First seen · 32 lines · 148 tokens per session scan A c8ca54ecc939
routinize is a skill published in the GitHub repository SoliEstre/EstreGenesis (8 stars, last pushed 4d ago), licensed Apache-2.0. It adds 148 tokens to every session and 1,027 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
specflo-auto
Use when the user wants an unattended specflo run — "auto mode", "autopilot", "run it on its own", "keep going without me", "don't stop to ask" — map that to specflo auto. It is an explicit, per-invocation opt-in that drives the pipeline across phase boundaries without the ask-first pause. Do NOT use for the normal…
discovery-question-form
Structured clarification form for unresolved material requirements.
cog-braindump-capture
Capture raw thoughts with automatic domain classification and vault routing.
cog-knowledge-consolidation
Build structured knowledge frameworks from scattered vault notes with source attribution.
context-engineering
Dynamic context injection, mode switching (dev/review/research), selective loading, and strategic compaction for token optimization.
todos
This chat has a shared, live TODO plan — your tasks for the conversation, which the user also edits. Read this skill and reach for the todo tools whenever a request takes more than a couple of steps. It covers the plan model (group = task, items = its steps; loose items are the user's lane), how to work it: propose…