memory

Manage SSOT, memory, and cross-tool memory search. Guardian of decisions.md and patterns.md. Use when user mentions memory, SSOT, decisions.md, patterns.md, merging, migration, SSOT promotion, sync memory, save learnings, memory search, harness-mem, past decisions, or record this. Do NOT load for: implementation work, reviews, ad-hoc notes, or in-session logging.

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

Quality

Content

72%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 well-structured, reference-driven skill body with strong progressive disclosure and concrete MCP tool listings. The main weakness is the dispatch workflow's lack of validation checkpoints for the destructive/batch operations (merging, migration, SSOT promotion) it routes to, which caps workflow clarity.

Suggestions

Add explicit validation checkpoints to the 実行手順 for destructive operations — e.g. after a plans.md merge or migration, verify the SSOT file still parses / decisions are not duplicated before considering the step complete.

Include one short inline worked example of a harness_mem_* call (e.g. a harness_mem_search invocation with expected arguments) so the core action is executable without opening a reference file.

Trim the Layer 1/2 explanatory sentence in the D22 section to just the promotion instruction ("`/memory ssot` で Layer 2 に昇格") to tighten conciseness.

Dimension	Reasoning	Score
Conciseness	Mostly lean and reference-driven: a compact feature table, short MCP tool lists, and brief dispatch steps with no padding of concepts Claude already knows. The "Claude Code 自動メモリとの関係（D22）" section explains the Layer 1/2 architecture a touch more than needed, keeping it just short of a 5.	4 / 5
Actionability	Concrete MCP tool names grouped by purpose ("検索: `harness_mem_search`, `harness_mem_timeline`...", "記録: `harness_mem_record_checkpoint`...") and clear per-feature reference pointers give mostly executable guidance. It stops short of a 5 only because core actions are delegated to references with no inline worked example of invoking a tool.	4 / 5
Workflow Clarity	The 実行手順 (1. classify request → 2. read the appropriate reference → 3. execute per its content) is a clear 3-step sequence, but it lacks validation checkpoints for destructive/batch operations (plans.md merge, migration, SSOT promotion), which caps workflow_clarity at 3 per the rubric.	3 / 5
Progressive Disclosure	The body is a clean overview whose feature table points one level deep to five real reference files (all present in ./references/), each clearly signaled via ${CLAUDE_SKILL_DIR}/references/... links; content is appropriately split and navigation is easy.	5 / 5
	Total	16 / 20 Passed

Description

95%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, third-person description that clearly states both capability and trigger conditions with rich natural keyword coverage and an explicit negative-scope clause for distinctiveness. The only soft spot is slightly generic action verbs in the "what" clause.

Dimension	Reasoning	Score
Specificity	Names the domain ("Manage SSOT, memory, and cross-tool memory search", "Guardian of decisions.md and patterns.md") and lists several concrete operations (search, merge, migrate, sync, promote) tied to concrete artifacts, with only minor abstraction in verbs like "manage"/"guardian". Not a 5 because the vividly concrete action list sits partly in the trigger clause and verbs are somewhat generic.	4 / 5
Completeness	Explicitly answers what ("Manage SSOT, memory, and cross-tool memory search. Guardian of decisions.md and patterns.md.") and when ("Use when user mentions...") with concrete trigger phrases, plus a "Do NOT load for" negative-scope clause that strengthens the when. Matches the anchor 5 example directly.	5 / 5
Trigger Term Quality	"Use when user mentions memory, SSOT, decisions.md, patterns.md, merging, migration, SSOT promotion, sync memory, save learnings, memory search, harness-mem, past decisions, or record this" gives comprehensive natural-term coverage with synonyms ("save learnings", "record this"), artifact names, and the tool name harness-mem.	5 / 5
Distinctiveness Conflict Risk	A clear niche (SSOT/decisions.md/patterns.md governance, harness-mem) with a "Do NOT load for: implementation work, reviews, ad-hoc notes, or in-session logging" exclusion clause that sharply reduces overlap with generic skills; minimal conflict risk.	5 / 5
	Total	19 / 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
relative_links	Relative link issues: 7 missing, 7 deeper-than-1-level, 1 suspicious	Warning

	Total	15 / 16 Passed

Repository: Chachamaru127/claude-code-harness
Path: opencode/skills/memory/SKILL.md
Commit: 7d8deba

Reviewed: about 15 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.