application-skills

application-skills provides a large catalog of app-specific “skills” (one per integration) for AI agents to perform actions (e.g., Gmail/Slack/HubSpot/Salesforce) using the Membrane platform and the Agent Skills specification. Skills describe each app’s structure, available actions, and how to invoke them via compatible agent CLIs/hosts.

Evaluated Mar 30, 2026 (0d ago)
Repo ↗ Automation agent-skills integrations oauth saas membrane automation claude-code-skill openclaw-skills cursor
⚙ Agent Friendliness
44
/ 100
Can an agent use this?
🔒 Security
57
/ 100
Is it safe for agents?
⚡ Reliability
18
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
0
Documentation
55
Error Messages
0
Auth Simplicity
85
Rate Limits
10

🔒 Security

TLS Enforcement
70
Auth Strength
75
Scope Granularity
30
Dep. Hygiene
40
Secret Handling
60

Based on the provided README, auth is managed by Membrane (suggesting tokens are abstracted away from prompts), but the README does not provide concrete evidence about TLS enforcement, secret logging/redaction, scope granularity, or dependency security posture.

⚡ Reliability

Uptime/SLA
0
Version Stability
30
Breaking Changes
20
Error Recovery
20
AF Security Reliability

Best When

You want broad, standardized app connectivity for an agent using the Agent Skills ecosystem, and you can rely on Membrane to handle OAuth/API-key flows and token refresh.

Avoid When

You need a single first-class API contract (REST/OpenAPI) for all integrations, or you require explicit, documented rate-limit headers/retry/idempotency behavior in this repo alone.

Use Cases

  • Enable an AI agent to perform app actions (send messages, create tickets, manage CRM records) without hardcoding per-app API logic
  • Rapidly expand an agent’s integration surface across thousands of SaaS apps
  • Provide standardized, app-specific action schemas so an agent can choose the right action and parameters
  • Use with agent runtimes that support the Agent Skills specification (e.g., OpenClaw, Cursor, Claude Code)

Not For

  • Direct low-level interaction with underlying third-party APIs (the repo is skill documentation/assets, not a general-purpose API gateway)
  • High-assurance, formally verified security workflows (claims about security are not evidenced in the provided README)
  • Use where you need strict guarantees about idempotency, rate-limit behavior, or retry semantics without consulting deeper skill/spec docs

Interface

REST API
No
GraphQL
No
gRPC
No
MCP Server
No
SDK
No
Webhooks
No

Authentication

Methods: OAuth (managed by Membrane) API keys (managed by Membrane) Token refresh (managed by Membrane)
OAuth: Yes Scopes: No

The README states auth is handled by Membrane (OAuth, API keys, token refresh), but does not detail scope model, consent flow, or how scopes are represented to the agent.

Pricing

Free tier: No
Requires CC: No

No pricing or free tier information is provided in the supplied README content.

Agent Metadata

Pagination
none
Idempotent
False
Retry Guidance
Not documented

Known Gotchas

  • Skill catalogs are only as reliable as the underlying per-app action schemas; without per-action docs you may need to validate parameters for each integration.
  • Because auth is delegated to Membrane, agent behavior and error handling may depend on Membrane’s credential state and refresh behavior rather than the agent runtime.
  • The README does not describe retry/idempotency semantics; agents should avoid assuming safe retries for create/update actions unless specified in the underlying skill/action docs.

Alternatives

Full Evaluation Report

Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for application-skills.

AI-powered analysis · PDF + markdown · Delivered within 30 minutes

$99

Package Brief

Quick verdict, integration guide, cost projections, gotchas with workarounds, and alternatives comparison.

Delivered within 10 minutes

$3

Score Monitoring

Get alerted when this package's AF, security, or reliability scores change significantly. Stay ahead of regressions.

Continuous monitoring

$3/mo

Scores are editorial opinions as of 2026-03-30.

6533
Packages Evaluated
19870
Need Evaluation
586
Need Re-evaluation
Community Powered