Dashboard Visibility Mirror
Cloud / InfraSets up an idempotent mirror symlink that connects a producer agent's output to the dashboard and returns a verified JSON result.
Live output preview
A plan is required to view this content
Choose a plan to access input format, sample outputs, and live previews.
View Plans →About the skill
Dashboard Visibility Mirror (Mirror Symlink)
A single-purpose infrastructure (plumbing) skill that makes producer agents' output visible on the dashboard. The dashboard scans only a single gate: agents/<slug>/data/outputs/<entity>/. The agent, however, writes its real output to a canonical path (e.g. ~/Desktop/Projeler/... or a custom folder specified by the user). This skill connects those two points with an idempotent symbolic link—if the link isn't set up, the outputs remain "invisible" on the dashboard.
When to use it
- When a producer agent generates files and wants to show them in the dashboard's
Raporlartab or the global/outputsfeed. - When the user's prompt mentions "write to this folder," "output path," "canonical path," or "mirror symlink."
- In PHASE 1 (determining the output path), run BEFORE the data-collection/writing phase (PHASE 2); if it fails, the run is aborted.
What it produces
A structured JSON result: ok, mirror_path, canonical_path, symlink_action (created / overwritten / already_correct), and verified fields. On error it returns code (INVALID_SLUG, PATH_TRAVERSAL, VERIFY_FAIL, etc.).
Important rules
It uses ln -sfn (never ln -s or cp -r); entity_slug is whitelisted (path-traversal protection); the setup is verified with readlink. Only Bash + Read are needed, no external API required. It contains no domain knowledge—pure infrastructure that works drop-in for all producer agents.
How do I use this skill?
Upload the dashboard-output-mirror.zip you downloaded as-is — no packaging needed, the format is already correct (folder at root).
- Open Settings → Customize → Skills
- Upload → select the
dashboard-output-mirror.zipyou downloaded - Claude reads
SKILL.md; the name + description appear. Ready ✅
Scripts run in Anthropic's code-execution environment (sandbox) — not on your machine.