Obot is an open-source platform that helps organizations manage, secure, and govern AI clients, models, tools, MCP servers, skills, credentials, and hosted workloads. It is for teams that need controlled access, identity and permissions, sandboxed execution, registries, and audit records across agents such as Claude Code, Codex, Cursor, and IDE tools. The catalogue entries supply skills and instructions that can be managed or used within the Obot platform.
Borrowing it
Nothing to install: this file belongs to obot-platform/obot. 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/obot-platform/obot/main/.claude/skills/run-obot/SKILL.mdgit clone --depth 1 https://github.com/obot-platform/obotWrote 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/obot-platform/obot/run-obot)<a href="https://agentmods.dev/skills/obot-platform/obot/run-obot"><img src="https://agentmods.dev/badge/skills/obot-platform/obot/run-obot/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/obot-platform/obot/run-obot"><img src="https://agentmods.dev/badge/skills/obot-platform/obot/run-obot.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 | $0.00057 | $0.00936 |
| Opus 5.5 | $0.00023 | $0.00374 |
| Sonnet 5.5 | $0.00011 | $0.00187 |
| Haiku 4.5 | $0.00006 | $0.00094 |
Grade A, and why
run-obot scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 24d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- Readiness: `curl -sf localhost:8080/api/healthz && curl -sf -o /dev/null localhost:5174`. How it starts
The opening of the file, as written. The whole thing — 48 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Verify a change in a running Obot
Launch
make dev # Go server on :8080 (dev mode, auth off) + SvelteKit UI on :5174, proxied through :8080
- Ready in seconds once Go is cached. If it isn't up within ~1 minute, read the log; don't keep waiting.
pnpmrequires Node^24 || >=26. If the default Node is outside that, run with a 24 on PATH, e.g.PATH=$(mise where node@24)/bin:$PATH make dev.- The UI must be on 5174. If an old Vite still holds 5174, the new one moves to 5175 and :8080 can't reach it. Kill leftovers by PID (
ss -ltnp | grep -E ':(8080|8443|5174) '), not withpkill -fpatterns that can match your own shell. - Readiness:
curl -sf localhost:8080/api/healthz && curl -sf -o /dev/null localhost:5174. - The first page load shows a "Welcome to Obot" EULA dialog; click Continue.
Optional: connecting a local MCP server through a vMCP
Only needed when the change involves connecting to MCP servers, vMCPs, or the MCP Tester. Skip this section otherwise.
Remote MCP servers on localhost are refused by default, so start Obot with OBOT_SERVER_DISALLOW_LOCALHOST_MCP=false. Then:
# catalog entry for a local streamable-HTTP MCP server
curl -s -X POST localhost:8080/api/mcp-catalogs/default/entries -H 'content-type: application/json' \
-d '{"name":"Test","runtime":"remote","remoteConfig":{"fixedURL":"http://localhost:9123/mcp/"}}'
# wrap it in a vMCP to connect to it (there is no POST /api/mcp-servers)
curl -s -X POST localhost:8080/api/vmcps -H 'content-type: application/json' \
-d '{"displayName":"Test","components":[{"name":"t","mcpCatalogID":"default","mcpServerCatalogEntryID":"<entry id>"}]}'
- MCP Tester (inspector) UI:
http://localhost:8080/mcp-servers/test/<vmcp id>→ Tools tab. - Protocol-level check:
POST localhost:8080/mcp-connect/<vmcp or vmcp instance id>with a JSON-RPCinitialize. - After adding a tool to a test server, an existing vMCP kept showing the old tool list, even after Refresh in the tester (cause not investigated). Creating a new catalog entry and vMCP (e.g. for the server on another port) showed the new tool.
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.
- 24d ago First seen · 48 lines · 57 tokens per session scan A 9c9b17216062
run-obot is a skill published in the GitHub repository obot-platform/obot (1,167 stars, last pushed yesterday), licensed MIT. It adds 57 tokens to every session and 936 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-14.
run-obot compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| run-obotobot-platform/obot | Skill | 1.2k | yesterday | MIT | A | 57 | Claude Code |
| octocode-chrome-devtoolsbgauryy/octocode | Skill | 957 | 2d ago | MIT | A | 87 | Codex |
| cometchat-js-v5-sdkcometchat/cometchat-skills | Skill | 162 | 22d ago | MIT | A | 142 | Claude Code |
| cometchat-angular-v5-pushcometchat/cometchat-skills | Skill | 162 | 22d ago | MIT | A | 84 | Claude Code |
| drive-mcpjam-playgroundMCPJam/inspector | Skill | 2.3k | today | none | A | 48 | Claude Code, Codex |
| fetch-linuxdo-postCastor6/tactus | Skill | 380 | 6mo ago | Apache-2.0 | A | 34 | Claude Code, Codex |
| opencliqqqqqf-q/Arkloop | Skill | 374 | 24d ago | none | C | 0 | Claude Code, Codex |