How do I install the Repo B Agent Bridge Safety skill?

Run npx skillmds@latest add grtninja/repo-b-agent-bridge-safety in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Repo B Agent Bridge Safety safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: makes network calls. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Repo B Agent Bridge Safety?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Repo B Agent Bridge Safety free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Repo B Agent Bridge Safety?

grtninja (@grtninja) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Repo B Agent Bridge Safety by grtninja

Name: Repo B Agent Bridge Safety
Rating: 4.2 (5 reviews)
Author: grtninja

Repo B Agent Bridge Safety

Operate and secure the Continue Agent Bridge in <PRIVATE_REPO_B>. Use when configuring bridge modes, validating /api/agent endpoints, enforcing controlled-write safety gates, or diagnosing bridge availability and permission failures.

grtninja Updated Sep 10, 2026

File contents

REPO_B Shim Agent Bridge Safety

Use this skill for Agent Bridge setup and safety enforcement.

Workflow

Configure bridge environment variables for session mode.
Validate capabilities endpoint and a read-only task.
Keep read_only as default; enable controlled_write only when required.
Enforce allow_write and apply gating rules.

Scope Boundary

Use this skill for /api/agent/* bridge-mode and write-safety controls.

Do not use this skill for:

MCP adapter or Comfy resource diagnostics lanes.
Windows-host vs WSL reachability checks.
General PR preflight/doc lockstep lanes.

Required Environment (PowerShell)

$env:REPO_B_CONTINUE_BRIDGE_ENABLED = "1"
$env:REPO_B_CONTINUE_BRIDGE_URL = "http://127.0.0.1:11420"
$env:REPO_B_CONTINUE_MODE = "read_only"
$env:REPO_B_CONTINUE_ALLOWED_ROOTS = "G:\GitHub\<PRIVATE_REPO_B>"

Optional controlled-write limits:

$env:REPO_B_CONTINUE_WRITE_ALLOWED_ROOTS = "G:\GitHub\<PRIVATE_REPO_B>\repo_b_repo_b_python_shim,G:\GitHub\<PRIVATE_REPO_B>\tests"
$env:REPO_B_CONTINUE_APPLY_ENABLED = "0"

Model-plane note:

Treat direct LM Studio http://127.0.0.1:1234/v1 as authoritative and select http://127.0.0.1:2337/v1 only as an explicit hosted alternative.
Treat http://127.0.0.1:9000 only as the non-model MX3 device/DFP/telemetry/RAG/support service boundary.

API Checks

Invoke-RestMethod http://127.0.0.1:9000/api/agent/capabilities | ConvertTo-Json -Depth 8

$body = @{
  task_type   = "analyze_files"
  prompt      = "Summarize required docs changes."
  paths       = @("README.md", "docs/PROJECT_SCOPE.md", "docs/SCOPE_TRACKER.md")
  allow_write = $false
  dry_run     = $true
} | ConvertTo-Json -Depth 8
Invoke-RestMethod -Method Post -Uri "http://127.0.0.1:9000/api/agent/tasks" -ContentType "application/json" -Body $body

Reference

references/bridge-safety-checklist.md

Loopback

If this lane is unresolved, blocked, or ambiguous:

Capture current evidence and failure context.
Route back through $skill-hub for chain recalculation.
Resume only after the updated chain returns a deterministic next step.

---name: repo-b-agent-bridge-safetydescription: Operate and secure the Continue Agent Bridge in <PRIVATE_REPO_B>. Use when configuring bridge modes, validating /api/agent endpoints, enforcing controlled-write safety gates, or diagnosing bridge availability and permission failures.---# REPO_B Shim Agent Bridge SafetyUse this skill for Agent Bridge setup and safety enforcement.## Workflow1. Configure bridge environment variables for session mode.2. Validate capabilities endpoint and a read-only task.3. Keep `read_only` as default; enable `controlled_write` only when required.4. Enforce `allow_write` and apply gating rules.## Scope BoundaryUse this skill for `/api/agent/*` bridge-mode and write-safety controls.Do not use this skill for:1. MCP adapter or Comfy resource diagnostics lanes.2. Windows-host vs WSL reachability checks.3. General PR preflight/doc lockstep lanes.## Required Environment (PowerShell)```powershell$env:REPO_B_CONTINUE_BRIDGE_ENABLED = "1"$env:REPO_B_CONTINUE_BRIDGE_URL = "http://127.0.0.1:11420"$env:REPO_B_CONTINUE_MODE = "read_only"$env:REPO_B_CONTINUE_ALLOWED_ROOTS = "G:\GitHub\<PRIVATE_REPO_B>"```Optional controlled-write limits:```powershell$env:REPO_B_CONTINUE_WRITE_ALLOWED_ROOTS = "G:\GitHub\<PRIVATE_REPO_B>\repo_b_repo_b_python_shim,G:\GitHub\<PRIVATE_REPO_B>\tests"$env:REPO_B_CONTINUE_APPLY_ENABLED = "0"```Model-plane note:- Treat direct LM Studio `http://127.0.0.1:1234/v1` as authoritative and select `http://127.0.0.1:2337/v1` only as an explicit hosted alternative.- Treat `http://127.0.0.1:9000` only as the non-model MX3 device/DFP/telemetry/RAG/support service boundary.## API Checks```powershellInvoke-RestMethod http://127.0.0.1:9000/api/agent/capabilities | ConvertTo-Json -Depth 8``````powershell$body = @{ task_type = "analyze_files" prompt = "Summarize required docs changes." paths = @("README.md", "docs/PROJECT_SCOPE.md", "docs/SCOPE_TRACKER.md") allow_write = $false dry_run = $true} | ConvertTo-Json -Depth 8Invoke-RestMethod -Method Post -Uri "http://127.0.0.1:9000/api/agent/tasks" -ContentType "application/json" -Body $body```## Reference- `references/bridge-safety-checklist.md`## LoopbackIf this lane is unresolved, blocked, or ambiguous:1. Capture current evidence and failure context.2. Route back through `$skill-hub` for chain recalculation.3. Resume only after the updated chain returns a deterministic next step.

Frequently asked questions