Borrowing it
Nothing to install: this file belongs to InDate/devharness. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/InDate/devharness/main/.claude/skills/instructing-sessions/SKILL.mdgit clone --depth 1 https://github.com/InDate/devharnessWrote 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/indate/devharness/instructing-sessions)<a href="https://agentmods.dev/skills/indate/devharness/instructing-sessions"><img src="https://agentmods.dev/badge/skills/indate/devharness/instructing-sessions/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/indate/devharness/instructing-sessions"><img src="https://agentmods.dev/badge/skills/indate/devharness/instructing-sessions.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 | $0.00098 | $0.01275 |
| Opus 5.5 | $0.00039 | $0.00510 |
| Sonnet 5.5 | $0.00020 | $0.00255 |
| Haiku 4.5 | $0.00010 | $0.00128 |
Grade A, and why
instructing-sessions 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 today.
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 — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Instructing sessions
What a session holds of devharness is the text devharness puts in front of it. Every character of that text is paid for twice: once when it arrives, and again on every later request of the session, since each request re-reads the whole context. A fact therefore earns its place by arriving at the moment the call it bears on is made, and by arriving there once.
The moments
A fact has one home: the channel that reaches the moment it is needed.
| Moment | What the agent needs then | Home |
|---|---|---|
| Start - the session begins | That devharness exists and which jobs to reach for it with | docs/mcp-instructions.md, plugin/hooks/session-start.mjs |
| Choice - a tool is being picked | What the tool is for | The tool description; the action enum carries the action names alone |
| First use - an action's first call in the session | How to use it well | A once-per-session block in that message's section of docs/messages.md |
| Error - a call failed or was refused | What was missing, and the call that recovers | The error text, with the recovering call written out |
| Worse path - the session does a job another way | That a devharness path exists and what it costs | A once-per-session block on the reply that sees the worse path |
Start text is paid by every session, used or not: it carries discovery and nothing else. A parameter's description carries what that parameter does, since the schema arrives at choice. A fact about one action of a tool arrives with that action: at its first use, or in its error. A fact needed at no moment has no home and is deleted.
Reference for people (design notes, catalogues, history) lives in docs/ and
the README, where no session pays for it.
Skills hold reference for people
A skill reaches a session only when the session loads it, and loads land late, rarely, or on a copy left behind by an earlier version. Text that a session needs to operate has to sit on a channel that reaches every session at its moment. A skill or reference holds what a person reads; an operational fact found there moves to its home in the table.
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.
- today First seen · 96 lines · 98 tokens per session scan A b69d9e58bda9
instructing-sessions is a skill published in the GitHub repository InDate/devharness (19 stars, last pushed today), licensed MIT. It adds 98 tokens to every session and 1,275 once invoked, about $0.0004 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 2026-10-30.
Other skills, from other repositories
browser-use
Direct browser control via CDP for web interaction: automation, scraping, testing, screenshots, and site/app work. Also covers hosted cloud browsers for AI agents: the browser-use Python library (Browser(usecloud=True)), Playwright, Puppeteer or any CDP client.
x402
Set up Browser Use Cloud payments with x402 — pay per request from a crypto wallet (USDC on Base mainnet), no signup or API key. Two setups it works out up front — "just use it" (set up a wallet so you or Claude Code can run cloud browser tasks paid from the wallet — Claude writes and runs throwaway scripts, nothing…
remote-browser
Controls an isolated Browser Use Cloud browser from a sandboxed machine with the current Browser Use CLI.
cloud
Documentation reference for using Browser Use Cloud — the hosted API and SDK for browser automation. Use this skill whenever the user needs help with the Cloud REST API (v2, v3, or v4), browser-use-sdk (Python or TypeScript), X-Browser-Use-API-Key authentication, cloud sessions, browser profiles, profile sync, CDP…
open-source
Documentation reference for writing Python code using the browser-use open-source library. Use this skill whenever the user needs help with Agent, Browser, or Tools configuration, is writing code that imports from browseruse, asks about @sandbox deployment, supported LLM models, Actor API, custom tools, lifecycle…
qa
QA-test a website or web app and return a 1-5 quality score (5 = flawless, 1 = broken) with evidence. Use when the user wants to test, QA, evaluate, score, or "check how good" a site, page, flow, or app — including a local dev server (e.g. "qa test localhost:5173", "does the checkout work?", "rate this landing page").…