Choose theme
AI Resources
TinyFish Skills
The TinyFish cookbook includes a reusable web-agent skill for searching the web, extracting page content, running longer agent tasks, and controlling browser sessions through TinyFish services.
The current skill documents four routes—search, fetch, agent, and browser—so readers can compare a quick lookup with extraction, a longer task, or direct browser control. Use this as a first read, not a recommendation. Open the original project before trusting details like terms, limits, privacy, cost, setup, or safety.
What it is
A reusable web-agent skill
The cookbook packages TinyFish web actions as instructions that another compatible agent setup can follow, rather than presenting a new base model or standalone assistant.
Why it stands out
Four routes for different web jobs
Search returns web results, fetch extracts a page, agent handles a longer goal, and browser gives more direct session control. That split helps readers choose only the access a task needs.
Availability
Public GitHub repository
The public skill documents TinyFish CLI commands and authentication. An API key is required. TinyFish currently lists Search and Fetch as free; Agent and Browser are metered services.
Why it matters
What makes it useful
An agent can find web pages, read their contents, or collect structured data from a site through one reusable skill. The cookbook supplies command examples for those jobs, so a builder has a starting point for adding web access to an existing agent.
What to know
Where it fits
The skill gives a coding agent instructions for calling TinyFish. The CLI and hosted services perform the web work; the surrounding agent still chooses the task and uses the returned results.
Notable points
What stands out
Reading a supplied URL and clicking through a website use different commands. The skill starts with Fetch for page content and moves to Agent when interaction is needed, such as navigating filters to collect product data. Browser supplies direct programmatic session control for a custom script.
Before using
What to review
The CLI installation and login or API-key setup documented in the skill.
Which TinyFish services the task calls and their current usage charges; public skill instructions do not make the hosted browser services local.
The Browser session creation command and returned connection details for direct programmatic control.
Which examples come from TinyFish itself and which are community contributions that the cookbook says are not official views or supported products.
Keep the API key and browser connection details out of public code and logs. Check which sites, accounts and logged-in sessions the agent is allowed to reach.
Choose page and session data suitable for sending to the hosted service, and check where returned data goes in the surrounding agent workflow. End browser sessions when the task is finished.
Reader fit
Who may find it relevant
Builders adding web research or website interactions to an existing coding agent.
People comfortable with a command-line tool and a hosted API account.
Readers looking for a standalone chat app would still need an agent around this skill.
Editorial note
Why LifeHubber lists it
A builder collecting product data for another agent step can request a JSON list, giving that step named fields to work with instead of only extracted page prose. The Agent route accepts a goal with a requested output, such as a JSON list of products. TinyFish's FAQ explains that a COMPLETED run status only means the automation finished without crashing; the returned result shows whether it collected the requested data. That distinction matters when connecting website extraction to an agent that will use the output in its next step.
Source links
Source materials
Reader note
Before relying on this entry
LifeHubber lists entries to help readers inspect AI projects, not to endorse them or prove they are safe, suitable, accurate, maintained, or right for a specific use. We do not verify every entry in depth. Before relying on anything listed, review the original materials, terms, privacy practices, limits, and risks that matter for your situation.
More in AI Agents
Keep browsing this category
Explore more AI agent projects.
Paperclip
paperclipai/paperclip
A self-hosted server and dashboard for coordinating agent teams through companies, goals, roles, issues, heartbeats, budgets, approvals, and persistent activity records.
Universal Modder
rehan-remade/universal-modder
Agent skills and a Python CLI for PC game modding, with engine research, generated assets, in-game checks, and shared field notes. Setup and compatibility depend on the game and workflow; fal generation requires a separate API key.
Nango
NangoHQ/nango
An agent and product integration platform for auth, OAuth, API proxying, TypeScript functions, MCP tool calling, observability, and external API access.
For project maintainers
Listed here? You can use the badge.
If you maintain a project with a current LifeHubber listing, you may add the optional “Listed on LifeHubber AI Resources” badge to its README, docs, or website. No introduction or permission request is needed.
