Skills that ship scripts
Every skill that ships scripts and what each declares it needs — packages, environment variables, hosts, programs and files — and how to invoke a skill from Claude Code or any other IDE.
Many skills ship helper scripts in a scripts/ folder, and each one declares what its scripts need in its own manifest.json: deps for packages, env for environment variables, and permissions for the hosts, commands and files the scripts touch. You can read that block before you install.
Most scripts run on the Python standard library. One needs a package: figma-diagrams (requests). Installing dependencies covers both.
Every skill with scripts
What each one declares in its manifest: the packages to install, the environment variables it reads, the hosts its scripts contact, the programs they run, and which files they touch (named-paths means only paths you pass in). The table is generated from the manifests.
| Skill | Pack | Packages | Environment | Network | Runs | Files |
|---|---|---|---|---|---|---|
agent-config-review | ai-engineering | — | — | — | git | named-paths |
agent-incident-response | ai-engineering | — | — | — | git | named-paths |
ai-usage-report | ai-engineering | — | — | — | — | named-paths |
backlog-triage | trackers | — | — | — | — | named-paths |
behaviour-equivalence | dev-team | — | — | — | — | named-paths |
brand-kit | design | — | — | — | — | none |
bundle-size-review | frontend | — | — | registry.npmjs.org | npx, npm, pnpm, yarn | named-paths |
cloud-cost-review | sre-oncall | — | — | — | — | named-paths |
codebase-tour | dev-team | — | — | — | — | none |
content-design | experience-design | — | — | — | — | none |
context-graph | core | — | — | — | git | named-paths |
cover-letter | career | — | — | — | — | named-paths |
delivery-metrics-report | sre-oncall | — | — | — | — | named-paths |
dependency-upgrade-review | dev-team | — | — | github.com, api.github.com, registry.npmjs.org, pypi.org, proxy.golang.org | git, gh, npm, pip, go | named-paths |
event-contract-draft | solution-architect | — | — | — | — | named-paths |
feedback-synthesis | product-manager | — | — | — | — | named-paths |
figma-diagrams | solution-architect | requests==2.34.2 | FIGMA_TOKEN | api.figma.com, pypi.org, files.pythonhosted.org | pip | named-paths |
file-to-markdown | converters | — | — | — | pdftotext, pdftoppm, tesseract | named-paths |
hypothesis-comparison | research | — | — | — | — | named-paths |
interactive-diagram | solution-architect | — | — | — | — | project |
judge-audit | ai-engineering | — | — | — | — | named-paths |
lean-canvas | product-manager | — | — | — | — | named-paths |
md-to-confluence | converters | — | CONFLUENCE_EMAIL (optional), CONFLUENCE_API_TOKEN (optional), CONFLUENCE_PAT (optional) | *.atlassian.net | — | named-paths |
md-to-docx | converters | — | — | — | — | named-paths |
md-to-pptx | converters | — | — | — | — | named-paths |
md-to-xlsx | converters | — | — | — | — | named-paths |
output-hygiene | core | — | — | — | — | named-paths |
pre-merge-review | dev-team | — | — | — | git, ruff, mypy, pytest, * | project |
rendering-strategy | frontend | — | — | — | — | named-paths |
resume-tailor | career | — | — | — | — | named-paths |
risk-register | grc | — | — | — | — | named-paths |
rollout-plan | dev-team | — | — | — | — | none |
skill-review | skill-engineering | — | — | — | — | named-paths |
slo-spec | sre-oncall | — | — | — | — | named-paths |
source-synthesis | research | — | — | — | — | named-paths |
sql-review | data-analytics | — | — | — | — | named-paths |
team-status-report | trackers | — | — | — | — | named-paths |
terraform-plan-review | infra-as-code | — | — | — | terraform | named-paths |
tracker-brief-sync | trackers | — | — | — | — | named-paths |
web-performance-review | frontend | — | — | registry.npmjs.org | lighthouse, npx | named-paths |
Invoke a skill
All skills are invoked in chat. Arguments are passed as plain text after the skill's trigger phrase (or via $ARGUMENTS when invoked as a slash command in Claude Code).
architecture-diagrams
Natural-language trigger (works in any IDE that has the skill installed):
Draw me a Mermaid diagram of a three-tier web app on AWS with an ALB, two ECS services, and an RDS Postgres backend.
Slash-command form (Claude Code):
/architecture-diagrams three-tier web app on AWS with ALB, two ECS services, RDS PostgresEverything after the slash command becomes $ARGUMENTS inside the skill.
figma-diagrams
Natural-language trigger:
Inspect the structure of this Figma file: https://figma.com/file/abc123/MyArchitecture
Slash-command form (Claude Code):
/figma-diagrams inspect https://figma.com/file/abc123/MyArchitecture
/figma-diagrams post-comment https://figma.com/file/abc123/MyArchitecture "Looks good — ship it."The skill parses $ARGUMENTS to figure out which Figma URL you mean and which action to take.