Name: mempalace-recall
Rating: 68.8 (1 reviews)
Author: MemPalace

mempalace-recall

Recall protocol for MemPalace — search the palace before answering about past work, people, projects, or prior decisions. Apply when the user asks what was decided, what happened before, who someone is, what was discussed last time, or anything that may already be filed in their memory palace; or when mempalace-recall is invoked. Complements the mempalace setup skill and requires the mempalace-mcp server.

Quality

83%

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

The canonical home for this skill is mempalace-recall in MemPalace/mempalace

Quality

Content

75%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.

A concise, actionable recall protocol with clear sequencing, a useful tool-selection table, and explicit error-recovery guidance. It would reach the top tier by adding full MCP tool-call signatures and hardening verification checkpoints.

Suggestions

Add a compact argument/signature reference for each mempalace_* tool (e.g. `mempalace_kg_query(entity, relation?, as_of?)`) so calls are copy-paste ready, not just named.

De-duplicate the supersede/invalidate/add guidance between the Protocol and Unhappy-paths sections — define once and reference it.

Tighten the Identity section or fold it into a single sentence; the persona preamble adds tokens without changing recall behavior.

Dimension	Reasoning	Score
Conciseness	Lean and well-organized with no padding over concepts Claude already knows; a few sentences (e.g. the Identity preamble, repeated supersede/invalidate guidance) could be tightened further. Not a 5 because of minor repetition between the Protocol and Unhappy-paths sections.	4 / 5
Actionability	Provides concrete tool names, a tool-selection table, and exact executable commands (`mempalace repair --mode from-sqlite --archive-existing --yes`). Not a 5 because MCP tool-call signatures are only partially specified (e.g. `mempalace_search` args are sketched but most tools are named without argument detail).	4 / 5
Workflow Clarity	Numbered Protocol gives a clear sequence with availability verification and explicit unhappy-path recovery branches. Not a 5 because the checkpoints are advisory rather than hard validate-then-proceed gates, and verification of recall correctness is implicit.	4 / 5
Progressive Disclosure	Well-sectioned body with one-level-deep, clearly signaled links to the canonical recall/coordination protocol docs in the repository. No bundle files exist locally, so structure is judged on the inline organization and external references, which are appropriately split with minor gaps.	4 / 5
	Total	16 / 20 Passed

Description

91%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.

A strong description that pairs a concrete statement of what the skill does with an explicit, natural-language trigger list. It distinguishes itself from the sibling setup skill and declares its server dependency, with only minor specificity and overlap concerns.

Dimension	Reasoning	Score
Specificity	Names the domain and several concrete actions — "search the palace before answering about past work, people, projects, or prior decisions" — with only minor coverage gaps (no mention of the specific search/supersede tools). Not a 5 because it stops at the high-level "search" action rather than enumerating the concrete recall operations the body covers.	4 / 5
Completeness	Clearly answers both "what" ("Recall protocol ... search the palace before answering") and "when" ("Apply when the user asks what was decided..."). Both are explicit and concrete, matching the top anchor.	5 / 5
Trigger Term Quality	Comprehensive natural trigger phrases users would actually say — "what was decided", "what happened before", "who someone is", "what was discussed last time" — plus the invocation name. Matches the top anchor for coverage of natural terms.	5 / 5
Distinctiveness Conflict Risk	Has a clear niche (recall vs. the complementary setup skill) and an explicit server dependency, with minor overlap risk against the sibling `mempalace` skill. Not a 5 because the two MemPalace skills share a name prefix and conceptual space.	4 / 5
	Total	18 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: MemPalace/mempalace
Path: .claude-plugin/skills/mempalace-recall/SKILL.md
Commit: 5ea40d3

Reviewed: 3 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.