Aise status

Skill jiayx01/PhysisAI/plugins/physisai/skills/aise-status

PhysisAI: trusted, local-first personal and project memory for Codex and Claude Code.

Install

npx -y skills add jiayx01/PhysisAI --skill aise-status

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

18 days oldThe repository was created 18 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

只读检查 PhysisAI 项目的候选状态、个人与项目记忆、待复审项、任务召回、宿主接入和事件链完整性。用户问当前学到了什么、哪些规则生效、是否健康、记忆是否膨胀或能否恢复时使用。

SKILL.md

2.3 KB, 442 tokens by cl100k_base, as published. Nobody here has run it

Inspect Evolution Status

Read-Only Checks

Run:

physisai verify --root <project-root>
physisai status --root <project-root>
physisai review --root <project-root>
physisai memories --root <project-root>

When the user asks what context a task would receive, use the zero-write task-context query:

physisai context --root <project-root> --query "<current task>" --json

physisai context merges user and project memory in real time and does not write an event or preview file. Treat its output as private.

If the task needs detail beyond that minimal output, read .ai-self-evolution/generated/index.md, search its summaries and triggers, and open only linked Markdown files. Do not scan events.jsonl as a retrieval corpus; use it only for integrity and recovery.

Inspect and report:

active adapters and user-store path;
user-layer versus project-layer memory counts and whether project knowledge stayed isolated;
candidate, validated, blocked, active, rejected, superseded, and rolled-back counts;
active memories and items marked review_required;
managed AGENTS.md / CLAUDE.md adapter blocks and private-state Git ignore status;
whether the fallback context, Markdown index, and non-empty layered detail files match the live user and project stores;
the index path, detail paths, and whether every active unexpired memory maps to the correct storage layer;
context size, skipped items, and whether retrieval triggers found relevant memories;
both event-chain integrity results and recoverability from their events.jsonl files;
doctor warnings for v0.1 project-global state or v0.2 role-memory compatibility;
stale, conflicting, duplicative, or low-utility items that need user review.

Do not promote, reject, roll back, migrate, edit, compact, or repair state from this skill. Generated Markdown is read-only and disposable; never treat a manual edit there as a memory change. If integrity verification fails, stop and report the exact failure before reading memory as authoritative.

What ships with it: 1 file

281 B alongside SKILL.md

agents/

openai.yaml281 B