Install
$ agentstack add skill-benja-pauls-serpentstack-find-skills Open-source listing, not yet scanned by AgentStack. Follow the source repository for install instructions.
Security review
⚠ Flagged1 finding(s); flagged for manual review. · v0.1.0 How review works →
- • Prompt-injection patterns
- • Secret / credential exfiltration
- • Dangerous shell & filesystem operations
- • Untrusted network calls
- • Known-malicious package signatures
- high Pipes remote content directly into a shell (remote code execution).
What it can access
- ● Network access Used
- ✓ 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.
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
Find & Create Skills
Discover community skills and create new ones when SerpentStack's built-in skills don't cover what the user needs.
When to Use This Skill
Use this when the user asks for something outside the scope of the built-in skills:
| Built-in Skill | Covers | |---|---| | scaffold | Generating new API endpoints and frontend pages following project conventions | | auth | Understanding auth, protecting routes, swapping providers | | test | Running and interpreting pytest and vitest | | db-migrate | Alembic migrations, schema changes, seed data | | dev-server | Starting, monitoring, and auto-fixing the dev environment | | deploy | Docker build, ECR push, Terraform apply to AWS | | git-workflow | Branches, commits, PRs, rebasing | | generate-skills | Interviewing a developer to produce project-specific skills for any codebase | | model-routing | Delegating code generation to on-device models (Ollama) to reduce costs |
If the user needs something else — monitoring, Stripe, email, Slack, S3 file uploads, PDF generation, browser automation, etc. — search for a community skill or official docs first before building from scratch.
Step 1: Search for Existing Skills
Via web search (recommended):
Search for: "SKILL.md" or site:github.com SKILL.md
Via package registries:
Many integrations have official SDKs with excellent docs:
# Python packages
uv search
# Check PyPI directly
# https://pypi.org/search/?q=
# NPM packages
npm search
Via official documentation:
For third-party services, always prefer the provider's official docs over community implementations — they're always up to date:
- Stripe: https://docs.stripe.com/api
- AWS S3: https://docs.aws.amazon.com/s3/
- SendGrid: https://docs.sendgrid.com/api-reference
- Twilio: https://www.twilio.com/docs/usage/api
Step 2: Evaluate Before Adopting
Before using any community skill or third-party integration, check:
- Read the instructions — does it actually do what the user needs?
- Check requirements — what dependencies does it need? Are those installed?
- Check for red flags — does it ask you to run
curl | bashfrom unknown URLs? Does it request API keys be sent to external services? - Check recency — is the source recently updated, or abandoned?
Step 3: Create a New Skill
If no existing skill covers what's needed, create one. A skill is a markdown file at .skills//SKILL.md.
Skill format:
---
name:
description: ""
---
#
## When to Use
## Steps
## Verification
Good skills are:
- Actionable — specific commands, real file paths, decision trees (not vague documentation)
- Self-contained — all the info needed to complete the task, without requiring external searches
- Linked to live docs — for third-party services, link to their official docs rather than hardcoding API signatures that go stale
- Adapted to project conventions — use
uv addnotpip install, put services inbackend/app/services/, use structured logging, etc.
Step 4: Adapt to SerpentStack Patterns
When following any external guide or community skill, adapt it to SerpentStack conventions:
- Python packages: use
uv addnotpip install - Config/secrets: add to
backend/app/config.pyas a Pydantic settings class, set via env vars in.env - Service layer: new integrations go in
backend/app/services/, not in route handlers - Route handlers: use
async defwithAsyncSession = Depends(get_db) - Frontend API calls: add typed functions in
frontend/src/api/, hooks infrontend/src/hooks/ - Logging: use
get_logger(__name__)with structured JSON events
Example: Adding Stripe Integration
User asks: "I want to add Stripe payments."
- Search: check Stripe's official Python SDK docs at https://docs.stripe.com/api
- Install:
cd backend && uv add stripe - Config: add
STRIPE_SECRET_KEYandSTRIPE_WEBHOOK_SECRETtobackend/app/config.py - Service: create
backend/app/services/stripe.pywith payment logic - Route: create
backend/app/routes/payments.pywith checkout and webhook endpoints - Create skill: write
.skills/payments/SKILL.mddocumenting the integration for future agent sessions - Verify: run
make testto ensure nothing broke
Skills Created by This Project
Add skills here as you create them.
| Skill | What it does | |---|---| | (none yet) | Run make dev and start building! |
Created a useful skill? Consider sharing it by opening a PR.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: Benja-Pauls
- Source: Benja-Pauls/SerpentStack
- License: MIT
- Homepage: https://www.npmjs.com/package/serpentstack
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.