memory-management

Use when the user asks to "remember project context", review saved findings, initialize runtime memory, archive stale work, reconcile notes, or erase a subject; manages authorized HOT/WARM/COLD working memory across all disciplines while preserving registry event ownership and privacy controls. Not for changing canonical registry facts - route those through the owning registry. 项目记忆/跨会话

Quality

81%

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

62%Weight 40%Scale 1-3

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

The body is well-structured with explicit workflows, validation gates, and concrete commands, but it is held back by repeated guardrail restatements and by several reference, script, and template paths that are cited but missing from the bundle. Tightening the duplication and fixing or removing the dangling links would raise both conciseness and progressive disclosure.

Suggestions

Consolidate the repeated registry-truth guardrail (never accept proposals / never write canonical facts) into the Skill Contract or Authority Order section and reference it from elsewhere instead of restating it in Capture, Consolidate, and Privacy.

Resolve the dangling references: either add the missing `state-model.md`, `registry-event-protocol.md`, `skill-contract.md`, `runtime-invocation.md`, `scripts/registry-events.py`, `scripts/validate-audit-artifact.py`, and `memory/templates/*` to the bundle or rewrite the links to point at files that actually exist.

For the commands depending on `AARON_SKILLS_ROOT`/registry scripts, note the install prerequisite inline or link to a setup reference so the guidance is copy-paste executable in a standalone install.

Dimension	Reasoning	Score
Conciseness	The body is dense and assumes Claude's competence without explaining basic concepts, but the registry-truth guardrail (never accept proposals / never write canonical facts) is restated across the Skill Contract, Authority Order, Capture, Consolidate, and Privacy sections, so it could be tightened; this matches the score-2 anchor of mostly efficient with some redundancy rather than the score-3 "every token earns its place."	2 / 3
Actionability	It provides specific, copy-paste-shaped commands (`registry-events.py init`, `is-suppressed <aggregate-id>`, `validate-audit-artifact.py`, concrete line/KB caps and age thresholds), but the referenced scripts and several reference files are absent from the bundle, so the commands are not fully executable as-is; this fits the score-2 anchor of concrete but incomplete guidance rather than fully copy-paste-ready.	2 / 3
Workflow Clarity	Seven numbered workflow sections are clearly sequenced with explicit validation checkpoints (validate-audit-artifact.py after preflight, "Confirm suppression by replay, not by a cached view", "Rebuild and verify projections") and a dedicated Decision Gates stop/ask section for destructive purge/erase operations, matching the score-3 anchor for clear sequences with feedback loops.	3 / 3
Progressive Disclosure	There is a clear overview and a Reference Materials section with four valid one-level-deep bundle links, but four parent-dir links (`../../references/state-model.md`, `../../references/registry-event-protocol.md`, `../../references/skill-contract.md`, `../../references/runtime-invocation.md`) and the referenced `scripts/` and `memory/templates/` paths do not exist in the bundle, breaking navigation; this matches the score-2 anchor of structure present but organization flawed rather than the score-3 easy-navigation bar.	2 / 3
	Total	9 / 12 Passed

Description

100%Weight 40%Scale 1-3

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 actions, uses natural trigger phrasings, answers both what and when explicitly, and defines a distinct niche with an explicit out-of-scope boundary. It is one of the higher-quality trigger descriptions.

Dimension	Reasoning	Score
Specificity	The description lists multiple concrete actions — "remember project context", "review saved findings", "initialize runtime memory", "archive stale work", "reconcile notes", "erase a subject" — matching the score-3 anchor for multiple specific concrete actions; it is not the score-2 level because the action coverage is comprehensive rather than partial.	3 / 3
Completeness	It explicitly answers both what ("manages authorized HOT/WARM/COLD working memory") and when via the "Use when the user asks to..." clause with explicit triggers, matching the score-3 anchor; it is not score 2 because the when guidance is explicit rather than implied.	3 / 3
Trigger Term Quality	It quotes natural user phrasings verbatim ("remember project context", "archive stale work", "erase a subject") that a user would actually say, matching the score-3 anchor for good coverage of natural terms; it exceeds score 2, which expects only some relevant keywords with missing variations.	3 / 3
Distinctiveness Conflict Risk	It carves a clear niche (HOT/WARM/COLD working memory plus registry event ownership) and adds an explicit exclusion ("Not for changing canonical registry facts - route those through the owning registry"), matching the score-3 anchor for a clear niche unlikely to conflict.	3 / 3
	Total	12 / 12 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_field	'metadata' should map string keys to string values	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 5 suspicious	Warning

	Total	13 / 16 Passed

Repository: aaron-he-zhu/aaron-marketing-skills
Path: protocol/memory-management/SKILL.md
Commit: 4a77c48

Reviewed: 9 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.