confirm-action
Ask the user to confirm before taking a significant action. Use before containment, remediation, or other impactful operations to ensure analyst approval. Presents options and waits for response.
63
54%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/confirm-action/SKILL.mdQuality
Discovery
67%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
The description adequately communicates both what the skill does and when to use it, with explicit trigger guidance. However, it is somewhat generic in its core capability (confirmation prompting) and relies on domain-specific jargon without covering natural language variations users might employ. The specificity of actions could be improved by listing more concrete scenarios.
Suggestions
Add more natural trigger terms users might say, such as 'approve action,' 'confirm before proceeding,' 'authorization check,' or 'safety prompt' to improve discoverability.
Increase specificity by listing concrete examples of significant actions requiring confirmation, e.g., 'blocking IPs, isolating hosts, deleting files, or modifying firewall rules.'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (significant actions requiring confirmation) and some actions (containment, remediation, impactful operations), but the core capability—asking for confirmation—is fairly generic and doesn't list multiple concrete distinct actions beyond 'presents options and waits for response.' | 2 / 3 |
Completeness | Clearly answers both 'what' (ask the user to confirm before taking a significant action, presents options and waits for response) and 'when' (before containment, remediation, or other impactful operations to ensure analyst approval), with explicit trigger guidance using 'Use before...' | 3 / 3 |
Trigger Term Quality | Includes some relevant terms like 'containment,' 'remediation,' and 'impactful operations,' but these are somewhat specialized security/IR jargon. Missing broader natural language variations a user might say, such as 'approve,' 'confirm action,' 'safety check,' or 'authorization.' | 2 / 3 |
Distinctiveness Conflict Risk | The concept of 'asking for confirmation before significant actions' is somewhat generic and could overlap with any skill that involves user interaction or approval workflows. However, the mention of containment and remediation narrows it to a security/incident response context, providing some distinctiveness. | 2 / 3 |
Total | 9 / 12 Passed |
Implementation
42%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill is well-organized and clearly communicates when confirmations should be used, with good examples of confirmation scenarios. However, it lacks actionable implementation details—Claude doesn't know *how* to present the question or collect the response (e.g., should it use a specific tool, format the message a certain way, or just ask in natural language?). The workflow steps are too abstract to be useful.
Suggestions
Add concrete implementation guidance: specify exactly how Claude should present the confirmation (e.g., a specific message format, tool call, or natural language pattern) rather than abstract steps like 'Display the question'.
Include a complete example showing the full interaction flow—what Claude actually says/does, what the user responds, and how Claude proceeds based on the response.
Add handling for unexpected responses (e.g., user says something not in the options list) and specify default behavior if the response is ambiguous.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is reasonably concise but includes some unnecessary structure for what is essentially a simple 'ask user and return response' pattern. The workflow steps (Present Question, Wait for Response, Return Response) are overly obvious and don't add value for Claude. The 'When to Use' section is helpful but could be more compact. | 2 / 3 |
Actionability | The skill lacks any concrete, executable code or commands. It describes a conceptual workflow with abstract steps like 'Display the question' and 'Collect the user's selection' without specifying how Claude should actually implement this. The examples show input/output formats but not how to execute the confirmation pattern in practice. | 1 / 3 |
Workflow Clarity | The three steps are listed but are trivially obvious (present question, wait, return). There's no guidance on what to do if the user provides an unexpected response, no timeout handling, and no validation of the response against the provided options. For a skill that gates destructive operations, the lack of error handling or fallback behavior is a gap. | 2 / 3 |
Progressive Disclosure | For a simple, single-purpose skill under 50 lines, the content is well-organized with clear sections (Inputs, Workflow, Outputs, When to Use, Examples). No external references are needed and the structure supports easy scanning. | 3 / 3 |
Total | 8 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
- Repository
- dandye/ai-runbooks
- Commit
086cbf6
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.