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 instructions/atman-33/workhub/claude-mdgit clone --depth 1 https://github.com/atman-33/workhubWrote 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/instructions/atman-33/workhub/claude-md)<a href="https://agentmods.dev/instructions/atman-33/workhub/claude-md"><img src="https://agentmods.dev/badge/instructions/atman-33/workhub/claude-md.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 | $0.01462 | $0.01462 |
| Opus 5 | $0.00731 | $0.00731 |
| Sonnet 5 | $0.00292 | $0.00292 |
| Haiku 4.5 | $0.00146 | $0.00146 |
Grade A, and why
workhub CLAUDE.md 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.
How it starts
The opening of the file, as written. The whole thing — 111 lines — stays where its author put it; the contents beside it link to each section on GitHub.
workhub
workhub — a Tauri 2 desktop "All-in-one Dev Hub" (Rust backend + React/TypeScript/shadcn-ui frontend), based on the devdeck codebase. Task management for humans and AI agents (Claude Code / OpenCode) backed by a dedicated Obsidian vault; repository management and a music player are planned as later modules.
Commands
npm install # frontend dependencies (first time)
npm run tauri dev # run the app with hot reload (or: npm run tauri:dev)
npx tauri build --no-bundle # release build -> src-tauri/target/release/workhub.exe
npm run build # typecheck + build frontend only
# in src-tauri/
cargo test # unit tests (on the local windows-gnu toolchain use
# `cargo test --release` — debug test exes fail to
# load with STATUS_ENTRYPOINT_NOT_FOUND)
cargo fmt # format (CI enforces --check)
cargo clippy -- -D warnings # lint (CI enforces)
Architecture
Backend (src-tauri/src/):
| Module | Responsibility |
|---|---|
lib.rs / main.rs |
Tauri builder, command registration |
commands.rs |
#[tauri::command] layer exposed to the frontend |
models.rs |
domain types (Task, Project, Settings, Config, ...) |
tasks.rs |
task Markdown parsing/writing, vault file watching, _ai/index/tasks.json |
vault_project.rs |
vault projects (projects/<slug>/): folder scan, layout findings, archive/restore, repo link |
git.rs |
git CLI integration (kept from devdeck for the repos module) |
actions.rs |
external launches (VS Code, terminal, AI agent with task context) |
storage.rs |
JSON persistence (~/.workhub/config.json; migrated from %APPDATA%\workhub\ — see storage::migrate_from_appdata) |
vault_settings.rs |
the vault-scoped settings split (<vault>/.workhub/settings.json); see .claude/rules/settings-placement.md before adding a setting |
ink/ |
screen-annotation overlay: Alt double-press gesture + transparent draw window (overlay.html / src/overlay/main.ts) |
rawkey.rs |
the single, shared Raw Input keyboard listener behind ink/ and clips/ — registration is per-process, so features register a consumer here instead of their own listener (see .claude/rules/global-keyboard-input.md) |
clips/ |
clibor-style snippet picker: bare double-tap Ctrl opens a popup (clips.html / src/clips-popup/) that pastes the picked snippet into the previously focused app |
paste.rs |
paste injection into the focused app (clipboard save/restore + synthesized Ctrl+V), shared by voice.rs and clips/ |
update.rs |
self-update against GitHub Releases (disabled by default; no releases yet) |
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.
- 4d ago First seen · 111 lines · 1,462 tokens per session scan A a7e6341a7db4
workhub CLAUDE.md is an instructions file published in the GitHub repository atman-33/workhub (2 stars, last pushed yesterday), licensed MIT. It adds 1,462 tokens to every session, about $0.0073 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 instructions, from other repositories
superbrain CLAUDE.md
Instructions for m3talux/superbrain, covering superbrain — repository rules, non-negotiables, public-facing content, versioning and workflow.
pioneer AGENTS.md
AGENTS.md instructions for pioneerdotai/pioneer, covering database architecture, runtime model, scheduling classes, never hold database capacity during other work and transactions and batches.
claude-bedrock CLAUDE.md
Instructions for iurykrieger/claude-bedrock, covering bedrock — claude.md, what is bedrock?, entity types, writing rules and language.
WeftCut AGENTS.md
AGENTS.md instructions for WeftCut/WeftCut, covering weftcut — agent instructions, agent skills, issue tracker, domain docs and testing.
llm-knowledge-bases CLAUDE.md
Instructions for rvk7895/llm-knowledge-bases, covering claude.md -- project instructions, project identity, configuration, skills and formatting rules (obsidian-compatible).
AgentDock AGENTS.md
Instructions for Cailiang/AgentDock, a project described as: Beginner-friendly desktop hub for AI coding clients, providers, Skills and MCP | 面向普通用户的 AI 编程客户端管理工具.