Features & Capabilities FAQs

Yes, our platform supports several users accessing the same agents – it's even something we recommend when re-engineering processes with the help of agents.

Access rights can be set per agent, starting from completely private agents to agents accessible to the entire team.

An AI Agent can be used like a normal AI chat, but can also be given a standing job to do.

Key Interface Sections

💬 Chat

Here you can chat with your agent, or discuss what its job should be.

📝 Instructions

These are standing instructions for your agent, describing how it should behave, how it should react to incoming events (such as emails), or if it should do recurring tasks. You can write the instructions yourself, or chat with the agent and then ask it to write the instructions.

📁 Docs

Here you can upload docs with additional context. The Agent will know about the existence of these documents and can read them when needed. You can also allow the agent to create its own documents.

⚡ Capabilities

Here you give the agent access to tools and integrations such as Slack, image generation, email send/receive, etc. Or just chat with the agent and it will suggest what it needs to do the job.

📊 Activities & Approvals

Here you can see a diary of what the agent has been up to. This is especially useful if the agent does stuff in the background.

⚙️ Settings

Here you can configure things like the agent's name and image, and advanced features like agent to agent communication.

🚀 Prompt Examples

Invoice Router Agent

"I receive a lot of emails with invoices that need to be routed to the right team. How can you help me with this, and what would you need from me?"

See Invoice Router demo video

Positive News Agent

"Check for positive feel-good news every day at 6pm and send me an email"

See Positive News Agent demo video

Company Research Agent

"Whenever I paste in a company name, I want you to do research on that company to find out if it fits our ideal customer profile. I'll upload a doc for you describing what we are looking for."

Next Steps

Start by chatting with your agent to explore its capabilities, or try one of the prompt examples above to see how agents can automate your workflows.

If your voice transcription is incorrectly translating English to Swedish (or another wrong language), this is a known issue where the underlying AI model sometimes guesses the wrong language during auto-detection.

Solution

Explicitly set the transcription language in your user profile preferences instead of relying on auto-detection:

1. Go to your user profile preferences
2. Find the voice/transcription settings section
3. Manually select "English" (or your preferred language) instead of "Auto-detect"
4. Save the settings

Why this happens

The AI model used for transcription sometimes makes incorrect assumptions about the language being spoken, especially when:

• There are mixed language environments
• The speaker has an accent
• Background noise affects detection
• The auto-detection algorithm makes an incorrect initial guess

Alternative approaches

If the issue persists after setting the language manually:

• Speak more clearly and distinctly
• Reduce background noise during voice input
• Consider restarting your session and trying again

Still having issues?

If manually setting the language doesn't resolve the transcription problems, contact our support team at support@abundly.ai with details about:

• Which language you're speaking
• Which language it's being transcribed as
• Your current language settings

This helps us identify if there are additional technical issues affecting your transcription experience.

You can invite team members to access your agents in two ways: at the team level or for individual agents.

Team-level access

1. Log into your Abundly account
2. Navigate to Members in the main menu
3. Add new team members who can then access agents based on their permissions

Agent-specific access

For more granular control over individual agents:

1. Select the specific agent you want to share
2. Go to Settings for that agent
3. Click on Team Access tab
4. Choose the appropriate access level:
   • Admin: Team members can do anything, including deleting the agent or changing access rules
   • Edit: Team members can interact with and modify the agent (instructions, capabilities, docs, etc.)
   • Use: Team members can interact with the agent but cannot make changes
   • None: Team members cannot see the agent exists

Best practices

• Start with 'Use' access for most team members to prevent accidental changes
• Grant 'Edit' access only to team members who need to modify agent behavior
• Reserve 'Admin' access for project owners or IT administrators
• Review access permissions regularly as your team and projects evolve

Getting started with team collaboration

Team access is particularly useful when re-engineering processes with AI agents, as multiple stakeholders can collaborate on the same agent while maintaining appropriate access controls.

The recommended approach for fine-tuning AI agents is to:

1. Chat directly with the agent and update instructions with more specific details
2. Be very explicit about requirements (companies, time periods, etc.)
3. Try this method for 1-2 attempts
4. If you don't see improvement after 1-2 tries, contact Abundly support for additional assistance

This iterative approach of refining instructions through direct interaction is generally effective for improving agent performance.

Common issues and solutions

Date handling problems: If your agent includes wrong dates or information from outside specified time periods, this is often caused by a known issue where Claude (the underlying language model) can "hallucinate" dates when they are not explicitly provided.

Current solutions:

• Be very explicit with date ranges in your instructions
• Double-check that today's date is clearly specified in agent instructions
• Abundly is working on a platform-level fix to automatically provide dates to agents

When to escalate

If you've tried updating instructions 1-2 times and still experience issues:

1. Contact Abundly support with specific examples of the problems
2. Include details about what the agent should vs. shouldn't include
3. Provide examples of incorrect outputs
4. Support can help identify technical issues or provide advanced troubleshooting

Best practices for effective instructions

• Be very explicit about criteria (specific companies, date ranges, content types)
• Include examples of what should and shouldn't be included
• Specify the current date clearly in instructions
• Test changes iteratively rather than making many changes at once
• Document what works and what doesn't for future reference