sekhar14/torus
Persistent research memory for agentic harnesses (Claude Code, Claude Desktop, Codex): a local knowledge graph of companies, sourced claims and investment theses, with SEC filing alerts and research workflows (onboard, challenge, memo, revi
ARCHETYPE
Skill Collector
Ten-plus skills loaded on demand. A procedural-knowledge library.
CONTEXT TAX · EVERY TURN
up to ~3.5k 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 torus -e TORUS_CAPTURE_MODE=YOUR_TORUS_CAPTURE_MODE -e TORUS_HARNESS=YOUR_TORUS_HARNESS -- torus serve 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 torus -e TORUS_CAPTURE_MODE=YOUR_TORUS_CAPTURE_MODE -e TORUS_HARNESS=YOUR_TORUS_HARNESS -- torus serve $ npx degit sekhar14/torus/plugins/claude-code/skills/audit .claude/skills/audit $ npx degit sekhar14/torus/plugins/claude-code/skills/challenge .claude/skills/challenge $ npx degit sekhar14/torus/plugins/claude-code/skills/distill .claude/skills/distill $ npx degit sekhar14/torus/plugins/claude-code/skills/explore .claude/skills/explore $ npx degit sekhar14/torus/plugins/claude-code/skills/ingest .claude/skills/ingest $ npx degit sekhar14/torus/plugins/claude-code/skills/memo .claude/skills/memo $ npx degit sekhar14/torus/plugins/claude-code/skills/onboard .claude/skills/onboard $ npx degit sekhar14/torus/plugins/claude-code/skills/research-protocol .claude/skills/research-protocol $ npx degit sekhar14/torus/plugins/claude-code/skills/review .claude/skills/review $ npx degit sekhar14/torus/plugins/claude-code/skills/scorecard .claude/skills/scorecard $ npx degit sekhar14/torus/plugins/claude-code/skills/status .claude/skills/status $ npx degit sekhar14/torus/plugins/claude-code/skills/undo .claude/skills/undo $ npx degit sekhar14/torus/plugins/claude-code/skills/whatif .claude/skills/whatif
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 (1)
| server | source | est. tokens |
|---|---|---|
| torus env: TORUS_CAPTURE_MODE, TORUS_HARNESS | binary | 2.5k |
Skills (13)
Similar rigs
dymchenkko/claude-study-agent
AI study agent for university courses, built on Claude Code. Learns from your own slides, PDFs and lecture recordings, quizzes you with spaced repetition, tracks mistakes, and builds mock exams.
Pragmatist 2.5k tok · · Q15
johnfkoo951/cmds-llm-wiki
LLM Wiki template — Karpathy 3-layer pattern + Gold In Gold Out purpose gate + dual Claude Code·Codex harness (11 commands, 2 hooks, 18 web clipper templates)
Skill Collector 10.4k tok · · Q20
bradtraversy/skillpass
A validated directory of AI skills to help your workflow. Post your skills and install others.
Skill Collector 6.1k tok · · Q5
rodrigomanhaes/flowyeah
Plan-to-PR pipeline for Claude Code.
Minimalist 2.1k tok · · Q8