Install
$ agentstack add skill-waitdeadai-minmaxing-visualizeworkflow ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ● Environment & secrets Used
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
/visualizeworkflow
Approval-first workflow for visual or experiential alignment. Plain /workflow remains autonomous; this command is the opt-in approval route.
New /visualizeworkflow runs do not mutate product code or root SPEC.md. When a saved visualization is approved and implementation continues, /opusworkflow is the default outer route and /visualizeworkflow is the inner contract. Direct continuation remains valid, but implementation must inherit the Claude/Opus planner-reviewer plus MiniMax-M2.7-highspeed executor policy.
outer_route: opusworkflow
inner_contract: visualizeworkflow
Modes
New Run
Use for:
/visualizeworkflow "build the dashboard"
Run the normal governed planning spine inline:
- Taste gate: read
taste.mdandtaste.vision; if missing, stop and route
to /tastebootstrap.
- Research brief: use the repo's effectiveness-first deepresearch protocol
when current facts matter; justify local-only when they do not.
- Code audit: inspect relevant repo anchors before planning.
- Introspection: run
pre-planhard-gate reasoning before freezing the draft. - Plan and Agent-Native Estimate.
- Draft
SPEC.mdinto the visualization run folder asdraft-SPEC.md. - Create the visualization package using the
/visualizecontract inline. - Stop before implementation.
Do not mutate product code or root SPEC.md during a new run. The draft spec stays under .taste/visualizations/{run}/draft-SPEC.md until approval.
Continue
Use for:
/visualizeworkflow --continue .taste/visualizations/YYYYMMDD-HHMMSS-slug
This means the user approves the saved visualization enough to implement.
- Read
visualization.md,draft-SPEC.md, andapproval.json. - Check live
git status, activeSPEC.md, and obvious repo drift since the
artifact was created.
- Archive a non-reused active
SPEC.mdbefore replacing it. - Promote
draft-SPEC.mdto rootSPEC.md. - Execute the implementation inline using
/workflowrules as the playbook,
but do not rely on nested custom-skill chaining.
- Verify against root
SPEC.md, run post-implementation introspection, archive
closeout, and summarize.
Revise
Use for:
/visualizeworkflow --revise .taste/visualizations/YYYYMMDD-HHMMSS-slug "make it denser"
Update visualization.md, draft-SPEC.md, and approval.json with the revision request. Stay stopped with WAITING_FOR_VISUAL_APPROVAL. Do not implement.
Run Folder
All new and revised artifacts live under ignored .taste/visualizations/:
.taste/visualizations/YYYYMMDD-HHMMSS-slug/
visualization.md
draft-SPEC.md
approval.json
prompt.txt # when an image prompt is produced
diagram.md|html|svg # when a schematic is produced
index.html # optional HTML companion for human review
image.png|webp|jpg # only when an image is actually generated
approval.json must be valid JSON:
{
"status": "WAITING_FOR_VISUAL_APPROVAL",
"task": "short task",
"surface": "frontend-web",
"mode": "image|prompt-only|markdown|Mermaid|HTML|SVG",
"root_spec_promoted": false,
"implementation_started": false,
"created_at": "ISO-8601 or unknown",
"updated_at": "ISO-8601 or unknown"
}
On continue, update status to APPROVED_FOR_IMPLEMENTATION before implementing and to IMPLEMENTED_VERIFIED after verification.
Approval Stop Output
Every new or revised run must end with:
## Visual Approval Required
- Status: WAITING_FOR_VISUAL_APPROVAL
- Run Folder: .taste/visualizations/...
- Visualization: .taste/visualizations/.../visualization.md
- Draft SPEC: .taste/visualizations/.../draft-SPEC.md
- Implementation Started: no
Approve and continue:
/visualizeworkflow --continue .taste/visualizations/...
Revise:
/visualizeworkflow --revise .taste/visualizations/... "requested change"
Hard Rules
/visualizeworkflowis the approval route;/workflowremains autonomous.- Never implement before
--continue. - Never claim an image exists unless an image artifact path exists.
- Backend, API, infra, and agent-runtime tasks may use diagrams or operational
narratives instead of fake UI mockups.
- Do not read or persist secrets, private customer artifacts, or
.envfiles. - HTML artifacts are companion review surfaces. Keep
draft-SPEC.md,
visualization.md, and approval.json as the canonical implementation and approval sources, and make the HTML point back to them.
- HTML must stay local and self-contained by default: no remote scripts, remote
stylesheets, remote fonts, analytics, absolute local paths, .. traversal, or secret-like material.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: waitdeadai
- Source: waitdeadai/minmaxing
- License: Apache-2.0
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.