AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Speckit Diagrams

skill-m0rtalphe0nix-dotfiles-speckit-diagrams · by M0rtalPhe0nix

Turn GitHub Spec Kit feature artifacts into editable Excalidraw diagrams and local previews. Use when the user asks to visualize, diagram, map, explain visually, or create an architecture, workflow, data model, story map, traceability map, or implementation roadmap from Spec Kit files such as spec.md, plan.md, research.md, data-model.md, contracts/, or tasks.md.

No reviews yet
0 installs
11 views
0.0% view→install

Install

$ agentstack add skill-m0rtalphe0nix-dotfiles-speckit-diagrams

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No 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 No
  • 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.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-m0rtalphe0nix-dotfiles-speckit-diagrams)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
26d ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

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 →
Are you the author of Speckit Diagrams? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Spec Kit Diagrams

Convert the current Spec Kit feature into a small set of decision-useful, editable diagrams by composing the installed excalidraw skill.

Dependency

Before generating anything, read ${CODEX_HOME:-$HOME/.codex}/skills/excalidraw/SKILL.md completely and read only the Excalidraw references it routes to for the selected layout. If the skill is absent, stop and ask the user to install fabricioartur/codex-excalidraw-skill.

Treat this skill's privacy and artifact-selection rules as overrides when they conflict with the base skill.

Resolve the Feature

  1. Work from the repository root containing .specify/.
  2. If the user names a feature directory, use it.
  3. Otherwise run .specify/scripts/bash/check-prerequisites.sh --json --paths-only and use FEATURE_DIR.
  4. If no current feature is resolved, inspect specs/. Use the only feature when exactly one exists; otherwise ask the user which feature to visualize.
  5. Never invent missing artifacts. State which available files informed each view.

Read the Artifacts

Read all relevant files in the resolved feature directory:

  • spec.md: actors, user journeys, priority, requirements, edge cases, and success criteria.
  • plan.md: architecture, technology choices, boundaries, constraints, and project structure.
  • research.md: decisions, alternatives, and rejected options.
  • data-model.md: entities, relationships, states, and validation rules.
  • contracts/: external interfaces, endpoints, events, schemas, and error behavior.
  • tasks.md: phases, dependencies, parallel work, and user-story coverage.

Use codebase inspection to verify named components and integrations when implementation already exists.

Choose Views

Generate only views supported by available information. When the user does not name a view, select up to three complementary diagrams:

  1. User journey from spec.md: actors, triggers, happy path, decisions, failure paths, and measurable outcome. Prefer swimlanes or a staged flow.
  2. System architecture from plan.md, contracts, and verified code: users/channels, application components, data stores, external services, trust boundaries, and major data flows. Prefer layered zones.
  3. Domain model from data-model.md: entities, cardinality, ownership, lifecycle states, and invariants. Prefer an ER-style layout.
  4. Requirement traceability from spec.md plus tasks.md: user stories and functional requirements mapped to implementation phases or task groups. Show missing coverage clearly.
  5. Implementation roadmap from tasks.md: dependency-ordered phases, parallelizable groups, milestones, and critical path. Prefer columns or a dependency graph.
  6. Decision map from research.md: decision, drivers, selected option, rejected alternatives, and consequences.

Split overloaded visuals instead of creating one unreadable canvas. Do not copy paragraphs into nodes; use short labels and preserve identifiers such as US1, FR-003, and T012 for traceability.

Generate Artifacts

  1. Create FEATURE_DIR/diagrams/ when output is requested.
  2. For each view, first create a compact JSON source at diagrams/-diagram-spec.json containing the title, source artifact paths, nodes, edges, groups, and any layout notes.
  3. Use the base Excalidraw skill to create:
  • diagrams/.excalidraw
  • diagrams/-preview.svg
  • diagrams/-preview.html
  1. Use generate_excalidraw.py for simple node-edge views. Use direct Excalidraw scene composition and the base skill's architecture-zone, swimlane, or composed-layout guidance when a generic graph would obscure meaning.
  2. Run validate_scene.py on every .excalidraw file and apply the base skill's visual quality gate.
  3. Visually inspect every SVG preview. Fix overlaps, clipped labels, ambiguous arrows, tiny text, and excessive density before delivery.

Privacy and Safety

  • Default to local-only generation. Pass --no-web-link or omit --web-link in every base script call.
  • Do not upload specifications, diagrams, financial information, code-derived architecture, or other project content to Excalidraw or any external service unless the user explicitly requests a shareable web link in the current turn.
  • If the user requests a web link, explain that the encrypted scene is stored on Excalidraw's public JSON service and that possession of the full URL grants decryption access.
  • Open preview HTML only when its SVG was generated from trusted project artifacts; do not embed or open arbitrary third-party SVG because the preview page inserts SVG markup directly.
  • Never call session_state.py clear.
  • Store persistent outputs only under the resolved feature's diagrams/ directory unless the user chooses another location.
  • Preserve the source artifact files; this skill is read-only with respect to spec.md, plan.md, data-model.md, contracts, and tasks.md.

Delivery

Show the preview images, link the editable .excalidraw files and source JSON files, and summarize:

  • which Spec Kit artifacts were used;
  • what each diagram communicates;
  • any ambiguity, contradiction, or missing coverage discovered while visualizing;
  • whether all outputs remained local or a web link was explicitly created.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.