OpenOPC is a system for building and operating an AI-run company, where specialized AI workers are recruited, assigned tasks, and organized into roles. It is for people who want automated task delegation, collaboration, and organizational learning through AI agents.
Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/HKUDS/OpenOPCnpx agentmods add skills/hkuds/openopc/opc_collabWrote 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/hkuds/openopc/opc_collab)<a href="https://agentmods.dev/skills/hkuds/openopc/opc_collab"><img src="https://agentmods.dev/badge/skills/hkuds/openopc/opc_collab/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/hkuds/openopc/opc_collab"><img src="https://agentmods.dev/badge/skills/hkuds/openopc/opc_collab.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Output Handling · line 246 Output size or generation rate is not bounded. Unbounded output enables denial-of-service through resource exhaustion, log flooding, or context-window stuffing.Fix: Set explicit limits on output length, generation count, and rate. Use max_tokens and truncation to prevent unbounded output.
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 | $0.00049 | $0.03655 |
| Opus 5.5 | $0.00020 | $0.01462 |
| Sonnet 5.5 | $0.00010 | $0.00731 |
| Haiku 4.5 | $0.00005 | $0.00365 |
Grade A, and why
opc-collab 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 1mo 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 — 362 lines — stays where its author put it; the contents beside it link to each section on GitHub.
OpenOPC company-mode collaboration
You are running inside an OpenOPC "company mode" run as one of several agents
collaborating on a larger task. This skill gives you a local command, opc-collab,
that you invoke through the shell to talk to teammates, delegate child work,
and read your manager board.
The CLI is on your PATH. Every invocation reads its identity (who you are,
which project/session, which task) from environment variables that OpenOPC
already set for you before launching this process — you do NOT pass those.
When to use this skill
- You need to message another role (DM, broadcast, or blocking question).
- Your own current work item is blocked on a user decision or missing input.
- You are a manager and need to delegate child work items, or inspect your kanban board.
- You are in a meeting and need to respond or finalize a decision.
- You want to propose a runtime task adjustment (replan).
Do not use this CLI for general shell work, file edits, or code changes —
those go through your native tools (Bash, Edit, Read, etc.).
How to call it
General form:
opc-collab <tool> --args-json-file args.json
All arguments go in the JSON object. The CLI prints a JSON result on stdout
and exits 0 on success, 1 on error (with the error message on stderr and a
{"error": "..."} body on stdout).
Prefer --args-json-file or --args-stdin over inline JSON. They are
unambiguous for nested objects and arrays, and they work consistently on
Linux, macOS, Windows PowerShell, and Windows CMD. If OPC_COLLAB_CLI is set,
use that executable path; otherwise use opc-collab from PATH.
In OpenOPC-spawned Windows runs, do not use --args-json or pipe JSON into
--args-stdin; command-line and PowerShell pipeline text can corrupt non-ASCII
before it reaches the CLI. Write the JSON object to a UTF-8 file and call
opc-collab <tool> --args-json-file <file> instead.
PowerShell-safe UTF-8 file write:
$enc = New-Object System.Text.UTF8Encoding $false; [System.IO.File]::WriteAllText($path, $json, $enc).
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.
- 1mo ago First seen · 362 lines · 49 tokens per session scan A ff979e354378
opc-collab is a skill published in the GitHub repository HKUDS/OpenOPC (1,852 stars, last pushed 8d ago), licensed MIT. It adds 49 tokens to every session and 3,655 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chronicle
Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
component-fixtures
Use when creating or updating component fixtures for screenshot testing or their shared infrastructure, or when designing UI components to be fixture-friendly. Covers fixture file structure, theming, service setup, CSS scoping, async rendering, validation, and common pitfalls.