personal-ai-operations-review

对已进入 OPERATIONS MODE 的 Personal AI / DSH 基础设施做只读运维复查：聚合 governance、durability、aic drift、proposal、memory/model/routing 与 personalization 行为监控（纠正率/重复纠正/选择泄漏），区分 NO ACTION、REVIEW、ACTION REQUIRED 与 EXTERNAL BLOCKER；用于日常健康检查、异常复盘、升级前检查，或用户询问“系统现在健康吗/需要管什么/AI 最近是不是又变难用了/个性化有没有退化”时。默认不修改 canonical，不为变绿而静默修复。

Quality

85%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

81%Weight 40%Scale 1-5

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

The body is a well-structured, mostly lean operations runbook with concrete commands, clear sequencing, explicit validation/classification checkpoints, and a read-only safety boundary. The main gap is that governance/proposal/durability detail is inlined rather than delegated to reference files, and a couple of script invocations lack exact paths.

Suggestions

Give exact invocation paths for capability_gov.py and static_gov.py (and the inbox proposal check) so every workflow step is copy-paste executable.

Move the detailed per-domain contracts (Evidence Freshness, Durability cause taxonomy, Overall Action Resolver) into a short referenced file, keeping SKILL.md as a lean overview with one-level-deep links.

Dimension	Reasoning	Score
Conciseness	The body is largely efficient and assumes Claude's competence (contracts, classification rules, command snippets without basic concept explanations), but the detailed contract sub-sections and example report add some length that could be trimmed slightly.	4 / 5
Actionability	It provides concrete executable commands (personal_status.py, aic validate, aic diff <target>, rpo_check.py --json, personalization_status.py --json) with specific flags and per-target coverage; minor gaps are that several referenced governance scripts (capability_gov.py, static_gov.py) are named without exact invocation paths.	4 / 5
Workflow Clarity	The numbered workflow is clearly sequenced with explicit evidence-state and cause-aware classification as validation checkpoints, an Overall Action Resolver that gates ACTION REQUIRED on current verified evidence, and a read-only safety boundary; destructive/batch handling is explicitly controlled.	5 / 5
Progressive Disclosure	Structure is good with well-labeled sections and a real one-level-deep bundle (scripts/personalization_status.py referenced by an executable path, and the file exists in ./scripts); however most detail is inlined in SKILL.md rather than split into separate reference files, so it is not a clean 5.	4 / 5
	Total	17 / 20 Passed

Description

88%Weight 40%Scale 1-5

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 is strong: it states concrete capabilities, explicit use-when triggers with natural user phrasings, and a distinct read-only operations niche. Third-person voice is maintained throughout, with no over-claims. Minor room for improvement in synonym/extension coverage and tightening the long sentence.

Dimension	Reasoning	Score
Specificity	The description lists multiple concrete operations (aggregate governance, durability, aic drift, proposal, memory/model/routing, personalization monitoring with correction rate/repeat correction/selection leakage) and concrete classification outcomes (NO ACTION/REVIEW/ACTION REQUIRED/EXTERNAL BLOCKER), giving comprehensive coverage of the skill's capabilities.	5 / 5
Completeness	It explicitly answers 'what' (read-only ops review aggregating the listed domains and classifying into the four action buckets) and 'when' (daily health checks, anomaly post-mortems, pre-upgrade checks, and the quoted user questions), both with concrete trigger phrases.	5 / 5
Trigger Term Quality	It includes several natural user-facing phrasings ('系统现在健康吗/需要管什么/AI 最近是不是又变难用了/个性化有没有退化') plus routine-use triggers; it misses some synonym variations and file extensions, but covers the most natural things a user would actually say.	4 / 5
Distinctiveness Conflict Risk	The OPERATIONS MODE scoping, read-only/canonical-mutation boundaries, and personalization behavior-monitoring niche make it clearly distinct from a building/design skill; minor overlap risk with general governance/durability skills keeps it just below 5.	4 / 5
	Total	18 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
referenced_paths_exist	Referenced path issues: 2 missing, 2 deeper-than-1-level	Warning

	Total	15 / 16 Passed

Repository: ooooooooooooooooooop/agent-tools
Path: skills/personal-ai-operations-review/SKILL.md
Commit: 3677cfc

Reviewed: ~8 hours 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.