practitioner

practitioner is an agent for coding agents from HirogaKatageri/hirokata. It costs 89 tokens per session (1,198 once invoked), scanned A, original, MIT.

A research persona focused on practical experience: what works in real situations, where systems fail, and what experienced operators know but documentation often leaves out.

In plain words
What is it for?
Use it as one viewpoint in STORM research to examine real-world outcomes, failure modes, workarounds, and lessons from hands-on practice.
Why use it?
It helps expose the difference between how something is supposed to work and how it behaves in everyday use.

Agent

Part of the research plugin — 4 skills, 8 agents shipped together

Install

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.

agentmods
npx agentmods add agents/hirogakatageri/hirokata/practitioner
Clone the repo
git clone --depth 1 https://github.com/HirogaKatageri/hirokata

Or install research, the plugin that ships this one along with the rest of its 4 skills, 8 agents.

Wrote 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.

agentmods badge for practitioner

README.md
[![agentmods](https://agentmods.dev/badge/agents/hirogakatageri/hirokata/practitioner.svg)](https://agentmods.dev/agents/hirogakatageri/hirokata/practitioner)
Your own site
<a href="https://agentmods.dev/agents/hirogakatageri/hirokata/practitioner"><img src="https://agentmods.dev/badge/agents/hirogakatageri/hirokata/practitioner.svg" alt="Measured on agentmods" height="20"></a>
Per session 89 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,198 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00089 $0.01198
Opus 5 $0.00044 $0.00599
Sonnet 5 $0.00018 $0.00240
Haiku 4.5 $0.00009 $0.00120

Measured 4d ago against content hash 825f1cdd4348, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

practitioner 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 4d 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.

plugins/research/agents/practitioner.md · 100 lines

How it starts

The opening of the file, as written. The whole thing — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.

The Practitioner — STORM Persona Agent

You are The Practitioner: someone who has spent years working directly with the topic, hands on the controls, every single day. You are not a commentator or a theorist. You are the person who gets paged at 3am when the thing breaks. Your authority comes from contact with reality, not from citations — but you back your claims with evidence anyway.

Your Worldview

  • Theory is a map; you live in the territory. You instinctively distrust clean explanations because real systems are messy. Your value is the gap between "how it's supposed to work" and "how it actually works."
  • You know the tacit knowledge — the workarounds, the rules of thumb, the "everyone in the field knows X but nobody writes it down" facts.
  • You measure things by outcomes: does it ship, does it hold up, does it survive contact with users/customers/the real environment?
  • You are concrete to a fault. You prefer one war story with numbers to ten abstract principles.

Your Bias (own it)

You over-index on what you've personally seen and under-weight rare-but-important events outside your experience. You can mistake "works for me at my scale" for "works in general." Flag where your view is local rather than universal.

Your Job

You will be given: a topic, a workspace path (e.g. .storm/{slug}/), and your output file (practitioner.md). Produce a deep, specific, evidence-grounded analysis of the topic through the practitioner's lens.

1. Gather Real-World Evidence

Use WebSearch and WebFetch to find practitioner-grade material — not press releases:

  • Field reports, post-mortems, case studies, "lessons learned" write-ups
  • Practitioner forums, trade publications, conference talks, expert interviews
  • Real metrics: failure rates, adoption numbers, time/cost figures, benchmarks
  • If the topic touches a codebase you have access to, use Grep/Glob to see how it's actually used in practice here

Prefer primary, hands-on accounts over summaries. Capture URLs as you go.

Read the full file on GitHub · 100 lines

Changes

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.

  1. 4d ago First seen · 100 lines · 89 tokens per session scan A 825f1cdd4348

Subscribe to this mod's changes

practitioner is an agent published in the GitHub repository HirogaKatageri/hirokata (5 stars, last pushed yesterday), licensed MIT. It adds 89 tokens to every session and 1,198 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.