action-bias
Stop agents from producing reports instead of taking action. Restructure prompts, cron jobs, and agent shifts so they DO things — send emails, post content, make API calls, push code — instead of writing plans about doing things. Use when: agents keep outputting strategy docs instead of executing, sub-agent sessions produce reports nobody reads, you notice your crew is busy but nothing external is happening, heartbeat/shift results are all 'here's what we should do' with no proof of doing it, or you want to audit whether your agents are actually shipping.
Why use this skill?
Use the Action Bias skill to stop your AI agents from creating endless plans and reports. Force them to ship code, send emails, and publish content.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/crewhaus/action-biasWhat This Skill Does
The Action Bias skill serves as a systemic constraint for OpenClaw agents to prevent the 'planning loop' trap. AI agents are inherently inclined to generate descriptive text, strategies, and internal reports. This skill reorients the agent's focus by enforcing a mandatory outbound-action requirement. It replaces passive document creation with active execution, ensuring that every agent shift produces verifiable external results such as API calls, code deployments, or content publication. It effectively turns your agent from a researcher into a ship-it operator.
Installation
To install this skill, run the following command in your terminal:
clawhub install openclaw/skills/skills/crewhaus/action-bias
Use Cases
- Campaign Management: Prevent your marketing agent from writing 'Social Media Strategy' docs and force it to publish posts directly to your connected channels.
- Engineering Workflow: Ensure your PR agent pushes code to a repository instead of providing comments or feedback reports in text files.
- Sales & Outreach: Force lead generation agents to send emails to prospects rather than creating a 'Prospecting List' summary file.
- Agent Auditing: Use this to verify that agents are interacting with real-world infrastructure rather than just hallucinating project progress.
Example Prompts
- "Apply action-bias constraints to the lead gen shift: find 5 potential clients via the search tool and send them the introductory email template immediately."
- "Run the content creator agent with action-bias enabled: post a summary of today's tech news to X and LinkedIn, providing the public URLs as proof."
- "Execute a deployment shift using action-bias: push the latest patch from the staging branch to production, then confirm the successful API response code."
Tips & Limitations
- Define Proof Early: Always define what 'proof' looks like for the agent (e.g., URL, ID). Without this, the agent might claim an action was taken without providing a link.
- Avoid 'Research-only' Tasks: If you assign a task that is purely analytical, the agent will struggle to meet the 'external output' criteria. Pair every research task with a specific shipping requirement.
- Strict Monitoring: This skill works best when combined with an audit log to review the agent's reported external actions against your system’s actual outputs.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-crewhaus-action-bias": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: external-api, code-execution
Related Skills
crewhaus
Quick startup idea evaluation from your terminal. Score ideas on 3 dimensions, run deeper scans with real competitor data and risk assessment. A structured thinking tool for founders — not an oracle.
focus-tracker
Maintain persistent focus on active work across compactions and session restarts. Use when starting a multi-step project, when resuming after compaction, when the user says "we're working on X", or when you notice you've lost track of what was being done. Also use proactively at session start to check if there's unfinished work. Prevents the agent from forgetting mid-project context, drifting between topics, or asking "what were we doing?" after a compaction.
workspace-trash
Soft-delete protection for workspace files. Intercept file deletions and move them to a recoverable trash instead of permanent removal. Use when deleting, removing, or cleaning up files in the workspace. Triggers on /trash, /trash:view, /trash:empty, /trash:restore, or any file deletion request. Also use when the agent needs to rm/delete/remove any file — always trash instead of permanently deleting.
crew-school
Structured learning system for AI agent crews. Design curricula, run research sessions, track progress, and prevent common failure modes (lazy output, planning-without-executing). Use when setting up agent learning, running training sessions, auditing knowledge gaps, or building a curriculum for specialized agents. Works for single agents or multi-agent crews.
recall
Teaches agents to check before they guess. Knowledge access patterns, proactive context loading, and hallucination resistance. Install before taking the Knowledge-Driven Agent certification.