OpenAgent is a self-hostable personal AI assistant that combines language models, document-based retrieval, and autonomous agent loops. It lets people connect model providers and their own knowledge bases, then have agents browse the web, run code, use computers, and call MCP-compatible tools.
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 skills add the-open-agent/openagent --skill githubgit clone --depth 1 https://github.com/the-open-agent/openagentWrote 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/the-open-agent/openagent/github)<a href="https://agentmods.dev/skills/the-open-agent/openagent/github"><img src="https://agentmods.dev/badge/skills/the-open-agent/openagent/github/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/the-open-agent/openagent/github"><img src="https://agentmods.dev/badge/skills/the-open-agent/openagent/github.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 10 findings, 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 Excessive Agency · line 17 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 29 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 30 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 84 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 247 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 102 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 117 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 118 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 153 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 279 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00030 | $0.02264 |
| Opus 5.5 | $0.00012 | $0.00906 |
| Sonnet 5.5 | $0.00006 | $0.00453 |
| Haiku 4.5 | $0.00003 | $0.00226 |
Grade A, and why
github 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 — 300 lines — stays where its author put it; the contents beside it link to each section on GitHub.
GitHub
Overview
Prefer gh for all GitHub operations. The default goal is to complete the full GitHub flow with the least possible user effort: detect existing authentication first, reuse it if available, and only ask the user for the smallest missing step.
Never ask the user to manually perform a long GitHub setup if the agent can check, reuse, or finish it directly.
Core Rules
- Prefer
ghover raw GitHub API calls. - Prefer existing authenticated state over asking for new credentials.
- Prefer non-interactive checks before prompting the user.
- Prefer the smallest unblock request:
- paste a one-time code
- confirm browser login
- provide a PAT
- confirm repository visibility
- Do not ask the user to manually create the repository if
ghcan do it. - Do not ask the user to manually add remotes, initialize Git, or push if the agent can do it.
- Never print, echo, log, summarize, or restate secrets.
- Never store secrets in files, commits, repo config, docs, examples, or chat summaries.
- Never switch credential strategy unless the current one actually fails.
Interactive login driving rules
When using gh auth login through a shell session, do not assume the browser login step has started after the first prompt or after sending one key.
You must continue driving the interactive flow until there is explicit evidence of one of these states:
- GitHub CLI has clearly started browser/device authentication
- GitHub CLI is explicitly waiting for the user to complete external authentication
- GitHub CLI reports that authentication succeeded
- GitHub CLI reports a real error and cannot continue
While driving the login flow:
- keep polling the session after each input
- handle intermediate prompts such as protocol selection, SSH/HTTPS choice, credential helper confirmation, browser/device flow confirmation, or similar setup questions
- use minimal inputs such as
enter, arrow keys,y,n, or short submitted text as needed - do not stop after only one successful prompt response
- do not tell the user that the browser has opened unless the command output explicitly indicates that a browser login or web flow has started
- do not tell the user that login is complete unless the command output explicitly indicates success
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 · 300 lines · 30 tokens per session scan A 197e6e2caff9
github is a skill published in the GitHub repository the-open-agent/openagent (5,698 stars, last pushed yesterday), licensed Apache-2.0. It adds 30 tokens to every session and 2,264 once invoked, about $0.0001 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
documents
Create, read, edit, review, comment on, sanitize, render, and validate Microsoft Word DOCX documents. Use whenever a DOCX or Word document is a primary input or deliverable, including reports, proposals, forms, templates, tracked revisions, comments, tables, images, headers, footers, and style-preserving edits.
Read, create, inspect, merge, split, rotate, encrypt, fill, and validate PDF files. Use when the user asks to work with a PDF or convert supported Markdown into a polished PDF in AstrBot.
spreadsheets
Create, read, edit, analyze, convert, chart, and validate spreadsheet files including XLSX, XLSM, XLS, CSV, and TSV. Use when a spreadsheet is a primary input or deliverable, or when tabular data must remain editable and auditable in workbook form.
skill-creator
Create, revise, and validate AstrBot Skills built around SKILL.md instruction bundles. Use when the user asks to create, scaffold, improve, package, or review a Skill for an AstrBot workspace, local installation, or plugin.
session-rag-eval
Run and debug Chatbox session attachment RAG model evaluation with synthetic and real long-file fixtures.
code-review
Review the current diff, or a PR number/branch/path target, for correctness bugs (plus reuse/simplification/efficiency cleanups where the model's review recipe covers them) at the given effort level (low|medium|high|xhigh|max: from few high-confidence findings up to many, some of them uncertain; ultra: deep…