prbe-ai/research-os-agent
Probe Research: experiment tracking for coding agents. Plugins and skills for Claude Code, Codex and Cursor plus a read-only MCP server, so agents build on what your team already tried. CLI: pip install probe-research / npx probe-research.
ARCHETYPE
Skill Collector
Ten-plus skills loaded on demand. A procedural-knowledge library.
CONTEXT TAX · EVERY TURN
up to ~5.8k tokens
Claude Code loads MCP tools on demand, so the real cost is usually lower.
Moderate · median rig: 2.3k · breakdown
Copy this rig
# review before running: this installs third-party code $ claude mcp add --transport http probe https://mcp.research.prbe.ai/mcp $ claude mcp add --transport http probe-research https://mcp.research.prbe.ai/mcp $ npx degit prbe-ai/research-os-agent/skills ./rig-research-os-agent/skills
MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.
$ claude mcp add --transport http probe https://mcp.research.prbe.ai/mcp $ claude mcp add --transport http probe-research https://mcp.research.prbe.ai/mcp
$ npx degit prbe-ai/research-os-agent/plugins/probe-research-pi/skills/artifact-design .claude/skills/artifact-design $ npx degit prbe-ai/research-os-agent/plugins/probe-research-pi/skills/artifact-diagramming .claude/skills/artifact-diagramming $ npx degit prbe-ai/research-os-agent/plugins/probe-research-pi/skills/audit-team-note .claude/skills/audit-team-note $ npx degit prbe-ai/research-os-agent/plugins/probe-research-pi/skills/edit-notes .claude/skills/edit-notes $ npx degit prbe-ai/research-os-agent/plugins/probe-research-pi/skills/instrument-code .claude/skills/instrument-code $ npx degit prbe-ai/research-os-agent/plugins/probe-research-pi/skills/notes-audit .claude/skills/notes-audit $ npx degit prbe-ai/research-os-agent/plugins/probe-research-pi/skills/overview-charts .claude/skills/overview-charts $ npx degit prbe-ai/research-os-agent/plugins/probe-research-pi/skills/overview-contract .claude/skills/overview-contract $ npx degit prbe-ai/research-os-agent/plugins/probe-research-pi/skills/overview-page .claude/skills/overview-page $ npx degit prbe-ai/research-os-agent/plugins/probe-research-pi/skills/overview-visuals .claude/skills/overview-visuals $ npx degit prbe-ai/research-os-agent/plugins/probe-research-pi/skills/plan-research-review .claude/skills/plan-research-review $ npx degit prbe-ai/research-os-agent/plugins/probe-research-pi/skills/probe .claude/skills/probe $ npx degit prbe-ai/research-os-agent/plugins/probe-research-pi/skills/track-work .claude/skills/track-work $ npx degit prbe-ai/research-os-agent/plugins/probe-research-pi/skills/visualize-progress .claude/skills/visualize-progress $ npx degit prbe-ai/research-os-agent/plugins/probe-research-pi/skills/write-pane .claude/skills/write-pane $ npx degit prbe-ai/research-os-agent/plugins/probe-research-pi/skills/write-preview .claude/skills/write-preview $ npx degit prbe-ai/research-os-agent/plugins/probe-research-pi/skills/write-summary .claude/skills/write-summary
This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 16.2k rigs:
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(./.env.*)",
"Read(**/.env)",
"Read(~/.ssh/**)",
"Bash(git push --force:*)",
"Read(**/*.pem)",
"Read(.env)",
"Read(~/.aws/**)",
"Read(**/*.key)",
"Read(**/.env.*)",
"Bash(git push --force*)",
"Bash(rm -rf:*)",
"Bash(rm -rf *)",
"Bash(sudo *)",
"Bash(sudo:*)",
"Bash(rm -rf /)",
"Bash(git push -f:*)",
"Read(.env.*)",
"Bash(git reset --hard*)",
"Bash(rm -rf /*)"
],
"ask": [
"Bash(git push:*)",
"Bash(git push *)",
"Bash(gh pr merge *)",
"Bash(git commit:*)",
"Bash(rm *)",
"Bash(gh pr merge:*)",
"Bash(gh pr create *)",
"Bash(git reset --hard:*)",
"Bash(npm publish:*)",
"Bash(git rebase *)"
]
}
} MCP servers (2)
| server | source | est. tokens |
|---|---|---|
| mcp.research.prbe.ai · "probe" | remote · remote | 2.5k |
| mcp.research.prbe.ai · "probe-research" | remote · remote | 2.5k |
Skills (17)
Similar rigs
prbe-ai/probe
Probe (useprobe.ai): experiment tracking for coding agents. Plugins and skills for Claude Code, Codex and Cursor plus a read-only MCP server, so agents build on what your team already tried. CLI: pip install probe-research / npx probe-resea
Skill Collector 3.3k tok · · Q0
jalilmtn/probe
Feature-level test health and change verification for coding agents. Local Studio, MCP, Claude Code and Codex plugins, with Android emulator support.
Pragmatist 2.6k tok · · Q0
Igor-lx/claudeHarness
Claude Code machine-enforced harness for AI coding agent: quality gates, automated code review, testing, validation, verification, guardrails and coding standards. A frontend development tool for AI-assisted code quality assurance that turn
Pragmatist 408 tok · · Q0
IsuruMaduranga/one-code
Claude Code with the model slot left open. The full workflow, open source, on any model or provider.
Pragmatist 1.2k tok · · Q3