Closed
Conversation
Remove built-in directory mounting functionality from the core codebase and convert it to an optional skill that users can apply to their forks. Changes: - Remove mount-security.ts (moved to skill resource) - Remove mount types from types.ts (AdditionalMount, MountAllowlist, AllowedRoot) - Remove validateAdditionalMounts from container-runner.ts - Remove MOUNT_ALLOWLIST_PATH from config.ts - Create .claude/skills/add-mount/ with: - SKILL.md: Step-by-step instructions for adding mount capability - mount-security.ts.resource: Full implementation reference This follows the skills-over-features philosophy where users explicitly opt into capabilities rather than having them bundled by default. https://claude.ai/code/session_01YY7LaFMGA7fEKSTfdTuy7J
Simplify the resource file naming convention. https://claude.ai/code/session_01YY7LaFMGA7fEKSTfdTuy7J
Directory mounting is now an optional skill rather than built-in. Updated all documentation to reflect this change: - docs/SECURITY.md: Mark mount security as optional, update diagrams - docs/SPEC.md: Update container config docs, architecture diagram - docs/REQUIREMENTS.md: Note skill requirement for additional mounts - groups/main/CLAUDE.md: Add note about /add-mount skill prerequisite - .claude/skills/setup/SKILL.md: Replace mount config section with pointer to /add-mount skill https://claude.ai/code/session_01YY7LaFMGA7fEKSTfdTuy7J
gavrielc
force-pushed
the
main
branch
2 times, most recently
from
c83bc27 to
80e68dc
Compare
Collaborator
|
@gavrielc Recommending close: targets pre-refactor file structure. mount-security code and container-runner have changed significantly since this was opened. |
omarzanji
approved these changes
Feb 14, 2026
omarzanji
left a comment
omarzanji
left a comment
There was a problem hiding this comment.
✅ Approved - Excellent Fix!
Problem Solved: Fixes the bug where all Discord agents appeared as "Omni" instead of their specific names (OmarOmni, PeytonOmni, etc.)
Root Cause Analysis: 💯
- Correctly identified that Discord handler runs in host process (not agent containers)
- Host only has one
ASSISTANT_NAMEenv var - All bots shared the same name when translating @mentions
Solution Quality: ✅
- Clean implementation: extracts agent name from group's
triggerfield - Proper fallback to
ASSISTANT_NAMEif group not found - Handles both DM and channel messages
- Only 10 additions, 1 deletion - surgical change
Code Review:
- ✅ Logic is sound
- ✅ Handles edge cases (trigger without @)
- ✅ Clear inline comments
- ✅ No security concerns
Suggestion for future: Consider adding a test case to verify agent-specific name extraction works correctly across multiple agents.
Impact: High - enables proper multi-agent coordination in Discord channels
Ship it! 🚀
This fixes QuarterPlan task: init-1771098984478-jjkah
jbaruch
pushed a commit
to jbaruch/nanoclaw-public
that referenced
this pull request
Apr 9, 2026
First of 11 migrations (issue qwibitai#35). Adds dedicated sync_tripit MCP tool and IPC handler with minimal env: only TRIPIT_*, RECLAIM_API_TOKEN, and GOOGLE_* (7 vars). No process.env spread. run_host_script still works for the remaining 10 scripts. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
jbaruch
pushed a commit
to jbaruch/nanoclaw-public
that referenced
this pull request
Apr 9, 2026
…bitai#35) Dedicated MCP tool + IPC handler with minimal env: TRAKT_CLIENT_ID, TRAKT_ACCESS_TOKEN (2 vars). No process.env spread. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
jbaruch
pushed a commit
to jbaruch/nanoclaw-public
that referenced
this pull request
Apr 9, 2026
…travel_schedule ref (3/11, qwibitai#35) check_travel_bookings already runs in-container via Bash — no host operation needed. refresh-travel-schedule.py also runs in-container (reads tripit-url.txt from disk). Updated SKILL.md reference. No more run_host_script references in any SKILL.md. Only 2 host operations remain: sync_tripit and fetch_trakt_history. The other 8 scripts all run in-container via Bash. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
kenansun-dev-bot Bot
pushed a commit
to kenansun-dev/nanoclaw-github-copilot
that referenced
this pull request
Apr 12, 2026
…wibitai#35) - nanoclaw init: copies container/skills/ to ~/.nanoclaw/skills/ - host-runner: uses package root (import.meta.url) instead of process.cwd() for skills fallback — works correctly after npm install 336/336 tests pass. Co-authored-by: Kenan Rpi5 Claw <rpi5-claw@nanoclaw.dev>
3 tasks
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Type of Change
Description
This PR reverts the directory mounting feature that was added in the
add-mountskill. The changes remove:AdditionalMount,MountAllowlist,AllowedRoot) fromsrc/types.tssrc/container-runner.tssrc/config.tsmount-security.tsmodule (via resource file removal)The skill documentation in
.claude/skills/add-mount/SKILL.mdremains as a reference for how to re-implement this feature in the future if needed.Rationale: This simplification removes unused mounting infrastructure from the codebase while preserving the skill documentation for future reference.
https://claude.ai/code/session_01YY7LaFMGA7fEKSTfdTuy7J