~ / rigs / atvirokodosprendimai / tool-multipathreadwrite

atvirokodosprendimai/tool-multipathreadwrite

Multi-path read and write for coding agents: many ranges, many edits, one call, a verdict per edit. Measured on SWE-bench Verified.

↗ GitHub ★ 3 MIT updated 13d ago project Claude CodeCodex
stack: Go
share on X
ARCHETYPE
Pragmatist
A balanced, no-drama setup: some rules, some tools, nothing extreme.
CONTEXT TAX · EVERY TURN
~17.4k tokens
Context hog · median rig: 2.3k · breakdown
GUARDRAILS
1/5
No YOLO mode · details
Quality 15/100 Guardrails v2

Copy this rig

# review before running: this installs third-party code
$ npx degit atvirokodosprendimai/tool-multipathreadwrite/.claude ./rig-tool-multipathreadwrite  # inspect, then merge into .claude/

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.

$ npx degit atvirokodosprendimai/tool-multipathreadwrite/.claude/skills/mrw .claude/skills/mrw
$ npx degit atvirokodosprendimai/tool-multipathreadwrite/.claude/skills/release .claude/skills/release
$ curl -fsSL --create-dirs -o .claude/agents/adr-challenger.md https://raw.githubusercontent.com/atvirokodosprendimai/tool-multipathreadwrite/main/.claude/agents/adr-challenger.md
$ curl -fsSL --create-dirs -o .claude/agents/go-implementer.md https://raw.githubusercontent.com/atvirokodosprendimai/tool-multipathreadwrite/main/.claude/agents/go-implementer.md
$ curl -fsSL --create-dirs -o .claude/agents/go-reviewer.md https://raw.githubusercontent.com/atvirokodosprendimai/tool-multipathreadwrite/main/.claude/agents/go-reviewer.md
$ curl -fsSL --create-dirs -o .claude/agents/scout.md https://raw.githubusercontent.com/atvirokodosprendimai/tool-multipathreadwrite/main/.claude/agents/scout.md

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 *)"
    ]
  }
}

Skills (2)

Subagents (4)

adr-challenger
model: opus
Read-only critique of a draft ADR and its tasks in tool-multipathreadwrite before it is accepted — the class audit, a fence that cannot go red, Out of Scope dispositions, staleness and boundary claims
go-implementer
model: sonnet
Implements a decided, bounded change in tool-multipathreadwrite through mrw — red test first, then the code, then the gates. Use for a task an Accepted ADR already specifies, or a fix a review named e
go-reviewer
model: opus
Read-only hostile correctness review of a diff in tool-multipathreadwrite — the boundary (ADR-007, ADR-071 junctions, ADR-077 state, ADR-081 device names), exit codes, receipts, all-or-nothing writes.
scout
model: haiku
Fast read-only lookups in tool-multipathreadwrite — where a symbol lives, which ADR governs a path, what a commit changed, what memoryd remembers. Use for a single-fact question, not for review or des

Hooks (2)

eventmatcherruns
PostToolUseBash|Write|mcp__mrw__mrw_read|mcp__mrw__mrw_writepython3 "${CLAUDE_PROJECT_DIR}/.claude/hooks/rules-on-read.py"
PostToolUseBashpython3 "${CLAUDE_PROJECT_DIR}/.claude/hooks/static-after-commit.py"

Similar rigs

copied ✓