Skip to content

AX-1992 - Document VS Code sandbox errors in README troubleshooting #76

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
MatanEden1 merged 3 commits into from
Sep 9, 2026
Merged

AX-1992 - Document VS Code sandbox errors in README troubleshooting #76

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
9a92c0d
AX-1992 - Document VS Code sandbox errors in README troubleshooting
MatanEden1 Sep 9, 2026
cb77060
AX-1992 - Name the sandbox bypass commands in README
MatanEden1 Sep 9, 2026
2ab6a01
AX-1992 - Drop 'Your JFrog setup is fine' from README
MatanEden1 Sep 9, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
11 changes: 11 additions & 0 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -236,6 +236,17 @@ The JFrog MCP is a remote HTTP server the plugin attaches at `https://${env:JFRO

If the agent keeps falling back to the `jf` CLI even though the JFrog tools are present, it usually means the tools aren't selected in the picker — enable them and start a new chat. Make sure `JFROG_PLATFORM_URL` is set in the IDE launch environment so `https://${env:JFROG_PLATFORM_URL}/mcp` resolves correctly.

### MCP management fails with `MODULE_NOT_FOUND`, `EROFS`, or "no jf CLI server(s) found"

VS Code's agent sandbox (`chat.agent.sandbox.enabled` set to `on`; macOS and Linux only) blocks the agent terminal from reading and writing under `$HOME` — where the skill's helper script, the npm cache, and your `jf` config live.

**Recommended — approve only the blocked command.** VS Code offers to re-run it outside the sandbox. That prompt repeats whatever command the agent just tried, so read it before accepting. Accept only if it looks like one of these, and deny anything else:

- `node .../jfrog-agent-guard-check.mjs`
- `npx --yes --registry <registry-url> @jfrog/agent-guard ...`

**Alternative — turn the sandbox off.** Set `chat.agent.sandbox.enabled` to `off`. Simpler, but it disables sandboxing for every agent command, not just these.

---

## Support
Expand Down
Loading