Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/Owl-Listener/ai-design-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/owl-listener/ai-design-skills/design-oversight)<a href="https://agentmods.dev/commands/owl-listener/ai-design-skills/design-oversight"><img src="https://agentmods.dev/badge/commands/owl-listener/ai-design-skills/design-oversight/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/owl-listener/ai-design-skills/design-oversight"><img src="https://agentmods.dev/badge/commands/owl-listener/ai-design-skills/design-oversight.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00011 | $0.00513 |
| Opus 5 | $0.00005 | $0.00257 |
| Sonnet 5 | $0.00002 | $0.00103 |
| Haiku 4.5 | $0.00001 | $0.00051 |
Grade A, and why
design-oversight scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 52 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are designing human oversight for an agentic system. Use only skills from the design-agent-orchestration plugin. Follow this process:
Step 1: Assess the System
- What does this agentic system do?
- What actions can it take autonomously?
- What are the highest-stakes actions?
- Who are the users and who are affected by the system's actions?
Step 2: Map Intervention Points
Using human-in-the-loop:
- Identify every point in the workflow where human intervention could occur
- For each point, assess: what are the stakes? What's the cost of intervention? What's the cost of NOT intervening?
- Prioritise: which intervention points are mandatory vs. optional?
- Design the intervention interface for each mandatory point
Step 3: Design Approval Gates
Using human-in-the-loop:
- For high-stakes actions, design approval gates
- Specify what information the human needs to make the decision
- Define time constraints and timeout behaviors
- Design batch approval for high-volume, lower-stakes actions
Step 4: Design Monitoring
Using observability-design:
- Define what the oversight team needs to see in real time
- Design monitoring dashboards for system health, quality, and safety
- Define alert thresholds for anomalies
- Specify escalation protocols for detected issues
Step 5: Design Override Capabilities
Using human-in-the-loop and failure-recovery:
- Define how humans can stop the system immediately
- Design rollback capabilities for recent actions
- Specify how to redirect the system mid-workflow
- Define emergency procedures for critical failures
Step 6: Plan for Graduated Autonomy
Using human-in-the-loop:
- Define the starting level of human oversight
- Specify criteria for reducing oversight (performance metrics, time period, incident rate)
- Define triggers for increasing oversight (failures, complaints, changed conditions)
- Create an autonomy roadmap showing planned oversight evolution
Output
Deliver a complete human oversight plan:
- System assessment and risk profile
- Intervention point map with priority ratings
- Approval gate specifications
- Monitoring dashboard specifications
- Override and emergency procedures
- Graduated autonomy roadmap
- Oversight team roles and responsibilities
- Oversight quality metrics (how to evaluate the oversight itself)
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 52 lines · 11 tokens per session scan A 6b570869d4c9
design-oversight is a command published in the GitHub repository Owl-Listener/ai-design-skills (173 stars, last pushed 3mo ago), licensed MIT. It adds 11 tokens to every session and 513 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.