alignment-checkpoint

Create, review, and validate an alignment checkpoint. Use when a request is ambiguous, high-stakes, multi-step, or requires explicit approval before tool use.

Quality

51%

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./Skills/agent-ops/alignment-checkpoint/SKILL.md

Quality

Content

35%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

This skill reads more like a set of abstract operating principles than a concrete, actionable skill. While it has reasonable structure and touches on important concepts like validation gates and progressive disclosure, it lacks the concrete examples, schemas, and executable guidance needed to reliably direct Claude's behavior. The workflow steps are too vague to be consistently followed without interpretation.

Suggestions

Add a concrete, complete example showing the full checkpoint flow: a sample ambiguous user request, the extracted goal, assumptions listed, success criteria, approach options presented, and the approval gate format—so Claude knows exactly what output to produce.

Define the output schema explicitly (e.g., a JSON or markdown template) for the checkpoint artifact, rather than just listing field names like 'goal extraction, assumptions, success criteria' without structure.

Make workflow steps concrete: replace 'Inspect 2-3 focused surfaces' with specific actions like 'Read the user's request, identify the primary goal, list 1-3 ambiguous elements, and present them as numbered questions.'

Add a concrete example of a validation gate and what 'stopping at a failed gate' looks like in practice (e.g., 'If the user responds with changes to assumptions, update the checkpoint and re-present before proceeding').

Dimension	Reasoning	Score
Conciseness	The skill is reasonably concise but includes some sections that are somewhat abstract and could be tightened. The Philosophy, Anti-Patterns, and some Workflow steps feel like general principles Claude already understands rather than novel, specific instructions. However, it avoids egregious verbosity.	2 / 3
Actionability	The content is almost entirely abstract guidance with no concrete code, commands, schemas, or executable examples. Phrases like 'Inspect 2-3 focused surfaces,' 'smallest action that advances the confirmed goal,' and 'narrowest real validator' are vague and not copy-paste actionable. The examples section provides dialogue snippets but no structured output format or concrete implementation steps.	1 / 3
Workflow Clarity	There is a numbered workflow with a validation section that mentions fail-fast and rerun-after-fix patterns, which is good. However, the steps are abstract ('Classify the requested mode,' 'Inspect 2-3 focused surfaces') without concrete checkpoints, specific validation commands, or clear decision criteria. The validation section mentions stopping at failed gates but doesn't define what a gate looks like concretely.	2 / 3
Progressive Disclosure	The skill references a deferred context directory under 'Infrastructure/references/deferred-skill-context/agent-ops-alignment-checkpoint/' and instructs loading only what's needed, which is good structure. However, no bundle files were provided to verify these references exist, and the main document itself could better separate its sections—the Inputs/Outputs sections hint at a schema but don't link to one.	2 / 3
	Total	7 / 12 Passed

Description

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 has a solid structure with an explicit 'Use when' clause that covers both what the skill does and when to use it. However, the core concept of an 'alignment checkpoint' is somewhat abstract, and the trigger conditions are broad enough to potentially conflict with other skills. The description would benefit from more concrete actions and more specific, natural trigger terms.

Suggestions

Add more concrete actions describing what an alignment checkpoint actually produces or involves (e.g., 'Generates a summary of understood requirements and asks for user confirmation before proceeding').

Include more natural trigger terms users might actually say, such as 'confirm understanding', 'check before proceeding', 'clarify requirements', or 'pause for approval'.

Dimension	Reasoning	Score
Specificity	Names the domain ('alignment checkpoint') and lists some actions ('create, review, and validate'), but the actions are somewhat abstract—it's unclear what an 'alignment checkpoint' concretely involves or what specific outputs are produced.	2 / 3
Completeness	Clearly answers both 'what' (create, review, and validate an alignment checkpoint) and 'when' (ambiguous, high-stakes, multi-step requests, or those requiring explicit approval before tool use) with an explicit 'Use when' clause.	3 / 3
Trigger Term Quality	Includes some useful trigger terms like 'ambiguous', 'high-stakes', 'multi-step', 'approval', and 'tool use', but these are fairly broad and may not match natural user language. Users are unlikely to say 'alignment checkpoint' directly, and common variations like 'confirm before proceeding', 'double-check', or 'clarify requirements' are missing.	2 / 3
Distinctiveness Conflict Risk	The concept of an 'alignment checkpoint' is somewhat niche, but the trigger conditions ('ambiguous', 'high-stakes', 'multi-step') are very broad and could overlap with many other skills that also deal with complex or sensitive tasks. The term 'approval before tool use' adds some distinctiveness but not enough to fully differentiate.	2 / 3
	Total	9 / 12 Passed

Validation

90%

Warnings & errors only

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
metadata_version	'metadata.version' is missing	Warning

	Total	10 / 11 Passed

Repository: jscraik/Agent-Skills
Commit: 8e7e19d

Reviewed: 5 days ago

Table of Contents

Discovery Implementation Validation

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.