tmpl/multi-agent-mux
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

refactor(skills): make skills portable across users/locations via workspace-relative paths + env var overrides

Browse Source 
Changes:
- skills/lib.sh:
  - HOME_DIR default changed from $HOME to <workspace_root> (workspace self-sufficient)
  - Added CLAUDE_PROJECT_DIR / LOCAL_BIN env var pattern (default $HOME, overridable)
- skills/tmux-agent-orchestrate-monitor/scripts/reconcile.sh:
  - STATE_DIR moved from $HOME/.cache/... to <workspace>/.cache/tmux-agent-orchestrate-monitor
- skills/tmux-agent-orchestrate-create/scripts/create_session.sh:
  - WRAPPER uses $LOCAL_BIN env var (default $HOME/.local/bin)
- 6 SKILL.md: examples and explanations updated to mention env var override capability

User/portability contract:
- Workspace-internal data: .hermes/ + .cache/ (moves with workspace)
- User/system data: $HOME/* (overridable via CLAUDE_PROJECT_DIR, LOCAL_BIN)
- All env vars follow: ${VAR:-default} pattern with documented defaults

Verified on isolated server -L agy-homeport-test (kill-server after):
- syntax check PASS
- E2E: defaults resolve to workspace-relative paths
- E2E: env var override correctly changes paths
- 0 leftover direct $HOME references in code
- Global skill non-interference verified
- Main isolated server -L multi-agent-canary untouched
This commit is contained in:
Godopu
2026-06-20 05:39:27 +00:00
parent ad7be264e7
commit cd9eec112d
10 changed files with 34 additions and 22 deletions
Download Patch File Download Diff File Expand all files Collapse all files
skills/tmux-agent-orchestrate-monitor/SKILL.md
+1 -1
View File
@@ -154,7 +154,7 @@ disk: ~/.claude/projects/.../87dc548e-...jsonl: missing
- **Don't run the monitor without `--goal`** — without goal mode, a single turn will spawn, do one reconcile, and complete. Goal mode keeps the worker alive across many turns.
- **The 30s poll is a default** — workers may override if they detect heavy churn. A workspace with 5+ agent sessions should bump to 60s to avoid noise.
- **`kanban_comment` rate limits** — Kanban may throttle if you comment too fast. Coalesce: only comment when the diff is *new* (not the same drift on every poll). The script tracks a state file at `~/.cache/tmux-agent-orchestrate-monitor/<workspace>.state` for this.
- **`kanban_comment` rate limits** — Kanban may throttle if you comment too fast. Coalesce: only comment when the diff is *new* (not the same drift on every poll). The script tracks a state file at `.cache/tmux-agent-orchestrate-monitor/<workspace>.state` in the workspace root for this (overridable via `AGENT_SESSIONS_STATE_DIR`).
- **Don't fight the user's explicit action** — if `tmux-agent-orchestrate-delete` is mid-flight and the monitor sees the same session in two states within 5s, prefer the user's most recent action. The monitor should not auto-revert a fresh `terminated` to `running` because of a stale `tmux has-session` check.
- **The monitor should never modify the conversation artifacts** (jsonl, db) — only the YAML. If you see a stale UUID, comment about it but don't delete the file.
- **TUI capture-pane is expensive** — only capture when you need to update `last_visible_status`, not every poll.