CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

memory-review

Review a project's archival_memory and propose promotions, near-duplicate merges, and stale cleanup for user approval. Read-only — applies nothing.

Quality

84%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

88%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 highly actionable, well-sequenced operational skill body with strong validation gates on the risky apply path. Its main improvement lever is moving long-tail operational detail (large-project tuning, apply mechanics, deferred features) into reference files and trimming duplicated safety statements.

Suggestions

Move the large-project performance notes (pool-cap explanation, --ef-search/--merge-timeout guidance, SQL prototype pointer) into a reference file and keep a two-line summary with the pointer in SKILL.md.

Split Phase-2a apply mechanics and the deferred-features list into a separate reference, leaving the overview to state that apply is dry-run-by-default and explicitly gated.

State the no-delete safety rule once in Rules and reference it from the apply section instead of repeating it.

DimensionReasoningScore

Conciseness

Dense and project-specific throughout — pool caps, phase limitations, learning-type routing — with essentially no filler; the body assumes Claude's competence. Not 5: the large-projects performance paragraph is a digression that could be tightened, and "never deletes the source archival row" is stated twice.

4 / 5

Actionability

Copy-paste-ready commands for the detector, dry-run, and execute paths with all flags documented, plus concrete judgment criteria (read the destination layer, check for prior capture, route ambiguity to the user). Not 4: the commands and decision guidance are complete, with no gaps to fill.

5 / 5

Workflow Clarity

Four clearly sequenced steps with explicit validation checkpoints for a database-touching workflow: dry-run first (always), verify resolved paths, confirm with the user, and a pg_dump backup that aborts the apply on failure, plus idempotent re-apply. Not 4: validation is explicit at every risky step rather than having minor gaps, so the destructive/batch cap does not apply.

5 / 5

Progressive Disclosure

Clear section structure (Usage, Process, Rules) with one-level references to the detector and apply scripts, but all ~115 lines live in SKILL.md. Not 5: operational detail such as the large-project tuning notes and Phase-2a apply mechanics could be split into a reference file to keep the overview lean.

4 / 5

Total

18

/

20

Passed

Description

75%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 specific, well-scoped description with concrete actions and strong distinctiveness, weakened only by the complete absence of trigger guidance. Adding an explicit "Use when..." clause would lift completeness and trigger quality simultaneously.

Suggestions

Add an explicit trigger clause, e.g. "Use when the user asks to review, tidy, or clean up project memory, promote learnings, or merge duplicate memories."

Include natural user-facing synonyms (consolidate, dedupe, prune, tidy) alongside the current technical vocabulary.

Mention the /memory-review invocation form so users who know the command name can find the skill by it.

DimensionReasoningScore

Specificity

Lists multiple specific concrete actions — "propose promotions, near-duplicate merges, and stale cleanup" — plus the "Read-only — applies nothing" safety constraint, which together cover the skill's full scope. Not 4: the action coverage is comprehensive rather than having minor gaps.

5 / 5

Completeness

The "what" is clear and specific (review archival_memory; propose promotions, merges, stale cleanup), but the "when" is entirely absent — no "Use when..." clause or equivalent trigger guidance, which caps completeness at 3 per the judging guidelines.

3 / 5

Trigger Term Quality

Good natural keyword coverage ("Review", "memory", "promotions", "near-duplicate merges", "stale cleanup") that users asking for memory curation would plausibly say. Not 5: misses common variations like "consolidate", "tidy", "prune", or "dedupe".

4 / 5

Distinctiveness Conflict Risk

"archival_memory" plus the promotion/merge/stale-cleanup vocabulary carves out a clear niche with minimal overlap risk against other skills. Not 4: the domain-specific terms make conflict unlikely even with closely related skills.

5 / 5

Total

17

/

20

Passed

Validation

93%

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

CriteriaDescriptionResult

referenced_paths_exist

Referenced path issues: 4 missing, 4 deeper-than-1-level

Warning

Total

15

/

16

Passed

Repository

stephenfeather/opc

Path

skills/memory-review/SKILL.md

Reviewed

Table of Contents

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.