review-alps
Skill1 starsChanged 24 days agoApache-2.0View on GitHub
Review ALPS repository changes for semantic consistency, source relationships, evaluation limits, and distribution integrity across the Process Framework, work-system design principles, design Skills, tools, examples, tests, and guidance. Repository-development Skill.
In this repository5 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- sync-localesSkill
- design-agent-work-systemSkill
- design-process-descriptionSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 6 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 25
Read the Framework and affected Process Descriptions in full.
- Distinctive: line 30
Follow required references and information relationships.
- Distinctive: line 31
Check that description validity, execution results, satisfaction of requirements, and their supporting evidence remain distinguishable.
- Distinctive: line 33
Check the distinct design and evaluation subjects of both Skills, the common source of target work meaning, and resource roles.
- Distinctive: line 39
Run the applicable format and repository integrity checks in the validation Workflow.
- Distinctive: line 44
Check specified tool behavior and necessary connections, including failures, incomplete results, side effects, and retry conditions where …
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
ALPS Repository Review
Purpose
Make inconsistencies and validation limits in an ALPS repository change clear enough to support a justified change decision.
Outcomes
- The change's effects on Process meaning and applicable requirements are identified.
- The consistency of affected sources, translations, and distribution resources is established or qualified by explicit findings.
- Findings and validation limits provide evidence for deciding how to proceed with the change.
Activities & Tasks
The following Tasks are required within the requested review scope.
Semantic assessment
- Identify the affected sources and inspect the complete task-owned diff, including additions and removals.
- Read the Framework and affected Process Descriptions in full.
- For work-system changes, read the design principles. Review responsibility allocation, exposed tool operations, information supply, execution conditions, and adaptation against the work. Verify the independent meaning of each foundation and its application in the relevant design Skill.
- Compare changed propositions by subject, modality, action or state, object, condition, quantifier, polarity, exception, and scope. Check that editorial changes preserve meaning and that intentional redefinitions have a basis in the requested design.
- Assess each description's Purpose and independently assessable Outcomes for relevance, individual necessity, and collective sufficiency. Check necessary detail, Activity/Task cohesion and coverage, boundary roles, and the distinction between work relationships and execution order.
- Review Markdown Process sections, Task lists, and note placement against the Framework, including
NOTEin English and注記in Japanese. Check that notes explain the relevant content while obligations remain explicit in the main text; assess meaning separately from formatting. - Follow required references and information relationships. Assess source identity, shared conditions, context-limited changes, and effects on related work.
- Check that description validity, execution results, satisfaction of requirements, and their supporting evidence remain distinguishable.
- Assess the basis for Process selection and changes, including necessary information from affected parties and the applicability of criteria and evidence. Check the effects of changed conditions on dependent judgments and the treatment of unresolved findings.
- Check the distinct design and evaluation subjects of both Skills, the common source of target work meaning, and resource roles. Assess the necessary correspondence between Tasks and tool operations, and between Process boundary information and tool interfaces.
- For rules or restrictions implemented by tools, examine their correspondence to applicable Controls and Constraints, their sources and scope, and evidence that implementation preserves them. Distinguish configuration limits from requirements of the work.
Distribution assessment
- Compare the distribution with AGENTS.md: distributed and development Skills, native Host manifests, required sources, and presentation resources.
- Run the applicable format and repository integrity checks in the validation Workflow. Examine path resolution from the Plugin's distributed layout.
- Apply sync-locales to affected English/Japanese pairs.
Implementation and work-system assessment
- Check specified tool behavior and necessary connections, including failures, incomplete results, side effects, and retry conditions where applicable.
- Assess whether the configuration, including rules and restrictions implemented by tools, respects applicable requirements, authority, and limits on effects, separately from specified tool behavior and effectiveness.
- Evaluate representative work using the intended agent, tools, information, and environment within the authorized scope. Relate results to the work's Outcomes and conditions, distinguishing this evidence from component tests.
- Identify the actual evaluation environment, evidence, unperformed checks, and limitations. Examine capability changes for effects on allocation, interfaces, and instructions while preserving work-specific criteria.
Finding synthesis
- Relate findings to their locations, evidence, effects, and coherent corrections.
- Report mechanical checks and semantic review separately, including failed or unperformed checks and unexamined scope. A passing format or path check does not establish Process meaning or Outcome achievement.
Controls
The Framework governs Process meaning and Markdown presentation, the work-system principles govern supporting configurations, and the design Skills apply these sources within their work. AGENTS governs repository layout.
Constraints
The review must remain within the requested change's scope. Respect the user's information and authorization boundaries. A review-only request returns findings; it does not authorize editing, executing the described work, or publishing changes.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

