virtuoso
Skill1k starsChanged 11 days agoMITView on GitHub
Operate Cadence Virtuoso through virtuoso-bridge: SKILL, CIW, schematic/symbol/layout, GDS, and ADE Maestro. Use spectre for standalone netlist-driven simulation.
Trending
More talk lately.thin dataabout the whole repository
Who's talking about Arcadia-1/virtuoso-bridge-lite
- cascode-ai/analog-canvas“Add modular Canvas export and native EDA import tools”GitHub · Oct 9, 2026
- sudosubin/agents.nix“agent-skills.github.arcadia-1.virtuoso-bridge-lite: 0.8.0 → 1.0.0”GitHub · Oct 5, 2026
- sudosubin/agents.nix“agent-skills.github.arcadia-1.virtuoso-bridge-lite: v0.8.0 → refs/tags/v0.8.0”GitHub · Oct 3, 2026
- sudosubin/agents.nix“agent-skills.github.arcadia-1.virtuoso-bridge-lite: init at 0.8.0”GitHub · Sep 25, 2026
The biggest 4 of 4 mentions linking to the repository that our crawl found. Each line is their own headline.
The file
Show its rules and text
We split this skill into the rules it gives an agent
13 rules. About 0 are common, 9 are distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈9
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Use the existing bridge for live Cadence operations.
- Very specific: line 20
Keep one-host use simple. Multiple servers/accounts/processes are optional.
- Distinctive: line 27
Use the existing project virtual environment and configuration.
- Distinctive: line 31
Verify SKILL function signatures against the target's installed Cadence docs and inspect actual library cells, symbol pins and CDF …
- Distinctive: line 33
Read local-docs.md for the verification workflow.
- Very specific: line 35
Check Python signatures against the active package, and confirm availability of version-specific commands with virtuoso-bridge --help.
- Distinctive: line 38
When an installed skill is a symlink, resolve its directory before following repository-relative paths.
- Very specific: line 44
Use create() for a new or explicitly authorized replacement cellview and modify() for an existing design.
- Distinctive: line 46
Read back connectivity and parameters, then check/save before netlisting.
- Distinctive: line 57
Before popup-prone operations or timeout recovery, read shared-ciw-dialogs.md.
- Distinctive: line 65
Do not resubmit, purge sessions or restart a pending daemon to unblock a wait.
- Distinctive: line 76
Read the branch needed for the operation.
- Very specific: line 95
When working in the repository, locate relevant examples with rg --files examples/01_virtuoso.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Virtuoso
Use the existing bridge for live Cadence operations. Python runs on the client; SKILL runs in the selected Virtuoso process. Local mode needs no SSH. Spectre execution is independent of this connection.
Choose the smallest interface
- One-off SKILL:
virtuoso-bridge eval 'EXPR',eval --stdin, orload FILE.il. - Common structured operations:
client.schematic,client.symbol,client.layout,client.libraryandclient.maestro. - Uncovered operations:
client.execute_skill(...)orclient.load_il(...). Raw SKILL is a supported escape hatch; a new wrapper is not a prerequisite.
Keep one-host use simple. Multiple servers/accounts/processes are optional; do not require profile inventory or process metadata for ordinary work. For GUI-equivalent simulation, preserve the netlist, models, corner and solver options and refresh the applicable PDK/CDF callbacks.
Before a live operation
Use the existing project virtual environment and configuration. If the bridge is not connected or the target changes, read connection.md. Pure documentation work and offline tests do not require a live connection.
Verify SKILL function signatures against the target's installed Cadence docs
and inspect actual library cells, symbol pins and CDF parameters before use.
Read local-docs.md for the verification workflow.
A miss in repository examples is not proof that a function does not exist.
Check Python signatures against the active package, and confirm availability
of version-specific commands with virtuoso-bridge --help.
When an installed skill is a symlink, resolve its directory before following repository-relative paths. References describe current supported behavior; an older client or daemon may lack the documented capabilities.
Editing and callback correctness
Use create() for a new or explicitly authorized replacement cellview and
modify() for an existing design. Preserve unrelated instances and user edits.
Read back connectivity and parameters, then check/save before netlisting.
Creation is not permission to delete an existing cell first.
For PDK devices, use the parameter/callback helpers where they support the
actual device; otherwise verify and invoke the required callbacks through SKILL.
Editable names and derived fields are PDK-specific: nf and fingers are not
universally interchangeable. Details:
schematic-skill-api.md.
Shared CIW and uncertain outcomes
Before popup-prone operations or timeout recovery, read shared-ciw-dialogs.md. It is the authority for dialog ownership, opt-in guards, request receipts and explicitly reviewed informational close.
Inspect through SSH/X11 when SKILL is blocked. Leave unknown-origin dialogs to the user; never automatically press Enter/Cancel or close the current form. Preserve the original request handle/history after uncertain execution. Do not resubmit, purge sessions or restart a pending daemon to unblock a wait. On old clients without the documented protection, use manual coordination rather than assuming that newer instructions make legacy execution safe.
Maestro lifecycle helpers can close/purge sessions. Use them only within the authorized lifecycle scope; on a shared CIW resolve the exact existing session. For results, bind reads/exports to the acknowledged history and use unique artifact paths. A sent key or nonempty output is not proof of a successful save/run.
Task references
Read the branch needed for the operation; do not load the entire catalog.
| Operation | Reference |
| CLI/Python calls, file transfer, batched attribute reads, CIW output | client-basics.md |
| Schematic editing, readback and CDF callbacks | schematic-python-api.md, schematic-skill-api.md |
| Recreate a schematic from an existing design | schematic-recreation.md |
| Import an exact-coordinate schematic manifest | schematic-manifest-import.md |
| Symbol generation/editing | symbol-python-api.md |
| Layout geometry, routing, vias and GDS operations | layout-python-api.md, layout-skill-api.md |
| Libraries and technology binding | library-python-api.md |
| SOS checkout/checkin of one explicitly selected cellview | sos-python-api.md |
| Maestro configuration, snapshots and result reading | maestro-python-api.md, maestro-skill-api.md |
| GUI-based simulation lifecycle | simulation-flow.md |
| Netlist import/export or batch netlisting without Maestro | netlist.md, batch-netlist-si.md |
| Cellview files, stale readback after another session saves, lock files and text/binary boundaries | cellview-on-disk-layout.md |
| SKILL Finder API | skill-finder-python-api.md |
| Errors, stalled batch tools or connection problems | troubleshooting.md |
When working in the repository, locate relevant examples with
rg --files examples/01_virtuoso. Use the corresponding API reference rather
than maintaining a second tutorial in this entrypoint.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

