Borrowing it
Nothing to install: this file belongs to honzapav/portuni. 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/honzapav/portuni/main/CLAUDE.mdgit clone --depth 1 https://github.com/honzapav/portuniWrote 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/honzapav/portuni/claude-md)<a href="https://agentmods.dev/instructions/honzapav/portuni/claude-md"><img src="https://agentmods.dev/badge/instructions/honzapav/portuni/claude-md/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/instructions/honzapav/portuni/claude-md"><img src="https://agentmods.dev/badge/instructions/honzapav/portuni/claude-md.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 | $0.06148 | $0.06148 |
| Opus 5.5 | $0.02459 | $0.02459 |
| Sonnet 5.5 | $0.01230 | $0.01230 |
| Haiku 4.5 | $0.00615 | $0.00615 |
Grade C, and why
portuni CLAUDE.md scanned grade C with 1 finding 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
rm -rf /Applications/Portuni.app How it starts
The opening of the file, as written. The whole thing — 401 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Portuni – Claude guide
Knowledge graph for organisations (POPP: organisations, projects, processes, areas, principles). Backend Node + libSQL (Turso), frontend React + Vite, desktop shell Tauri 2.
Workspace rule, before anything else. A workspace is either a team
workspace or a personal workspace, named by what it is for, not by
where its database sits. The team workspace is the primary one: a team runs
the central server, every teammate's desktop is a team workspace whose
sidecar is a sync agent, and that is where every real task, mirror and MCP
session happens. A personal workspace is the same server in a box for one
person; it must keep working, but it is not the reference. Every change to
the server, a REST route, an MCP tool or the session runtime works in
both before its issue closes. A half that is missing is an open issue
named in the PR title, never a "known gap" note in the docs. New behaviour
is written once as domain code that runs on the central server and in the
sidecar; where the device lacks the graph db, it takes a CentralClient
seam, not a second implementation. Words kept apart: team workspace
(data_mode: "central"), personal workspace (data_mode: "local"), the
central server (the process at api.portuni.com), the sync agent
(the device's sidecar in a team workspace, PORTUNI_AGENT_MODE=1) and
device-local (what the sync agent serves itself, device_local in code).
Model and checklist: docs/architecture/data-modes.md.
Where the rules live
Detailed behaviour and invariants are in docs/architecture/; this file keeps
the workflow and one-line rules with pointers. Read the doc before touching
its area.
| Doc | Read when touching |
|---|---|
data-modes.md |
anything that behaves differently on the central server, on a team-workspace device, or in a personal workspace; request routing; the workspace checklist |
file-state-and-sync-runs.md |
mirrors, watcher, reconcile, sync runs and jobs, remote watcher, locking, delete/move/rename, repair_needed |
file-sync.md, file-mutation-propagation.md |
the file-bytes plane design, adapters, tombstones |
sessions-and-runner.md |
sessions, tasks, runs, the Claude adapter, live channel server side, provider instances |
task-surface-web.md |
Práce, SessionChat, sidebar threads, the sessions client, AI Elements |
desktop-shell.md |
config.json, workspaces, windows, quit, events, localStorage, api_request routing, the WS bridge, updates |
mcp-scope-and-integrations.md |
MCP connect and scope configs, disk read scope, orientation, elicitation deadlines, Showtime, Vibe |
database-and-dialects.md |
schema, migrations, dialect-neutral SQL, tests on both drivers, export/import |
docs/env-vars.md |
any process.env read (PORTUNI_ROOT is write-scope tiering, PORTUNI_WORKSPACE_ROOT is mirrors) |
docs/lessons-learned.md §7 |
before every migration |
docs/vision/portuni-as-workspace.md |
product direction, "Local vs. central" 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.
- yesterday Changed · +11 lines · +173 tokens per session 53f79e29dbd8
- 28d ago Changed · -3 lines · -79 tokens per session 5f23914d0467
- 1mo ago Changed · +14 lines · +245 tokens per session 260b6ec758ac
- 1mo ago Changed · -2 tokens per session 499a7cfce675
- 1mo ago Changed · +9 lines · +165 tokens per session b753228c5691
- 1mo ago Changed · +6 lines · +144 tokens per session 768909803642
- 1mo ago Changed · +29 lines · +482 tokens per session 6be27e1c28c9
- 1mo ago Changed · +3 lines · +42 tokens per session 22992f85d6eb
- 1mo ago Changed · -1,994 lines · -36,774 tokens per session scan D → C 4fafd57442bf
- 1mo ago First seen · 2,326 lines · 41,752 tokens per session scan D 369decf8db82
portuni CLAUDE.md is an instructions file published in the GitHub repository honzapav/portuni (2 stars, last pushed yesterday), licensed Apache-2.0. It adds 6,148 tokens to every session, about $0.0246 per session on Opus 5.5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
portuni CLAUDE.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| portuni CLAUDE.mdhonzapav/portuni | Instructions file | 2 | yesterday | Apache-2.0 | C | 6,148 | Claude Code |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | yesterday | MIT | A | 7,461 | Codex, OpenCode |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | yesterday | MIT | A | 4,680 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 5,516 | GitHub Copilot |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 3,093 | GitHub Copilot |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 274k | 2d ago | MIT | A | 4,313 | Codex, OpenCode |