inland-empire

Subconscious memory layer. Absorbs observations, surfaces associative memories as hypotheses, and builds context across sessions. Commands: remember, consult, surface, forget, stats.

Quality

65%

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

Fix and improve this skill with Tessl

tessl review fix ./inland-empire/SKILL.md

Quality

Content

65%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.

The body is highly actionable with executable commands and concrete responses, but it carries metaphorical padding, lacks an explicit verify-then-delete workflow for the destructive forget command, and inlines all detail rather than using reference files despite a sizable bundle.

Suggestions

Trim the philosophical framing (the opening quote, "raw data of the soul", the "two registers"/"filing cabinet" prose) to reduce token padding without losing the consult-vs-surface distinction.

Turn `forget` into an explicit validate-then-delete workflow: instruct running `forget --dry-run` first, reviewing the candidate list, then re-running without --dry-run, so workflow_clarity can exceed the destructive-operation cap.

Offload the Storage Architecture and Configuration detail into a reference file (e.g. references/ARCHITECTURE.md) and link to it one level deep, surfacing the bundled backend directories as navigable references.

Dimension	Reasoning	Score
Conciseness	Mostly efficient with concrete commands and JSON examples, but the metaphorical framing ("raw data of the soul", "gut feeling that interrupts you mid-thought", the "two registers" prose) adds padding Claude does not need.	3 / 5
Actionability	Every command has copy-paste-ready invocations ("python3 scripts/inland_empire.py remember ..."), realistic options (--depth, --type, --before, --dry-run), a type-inference table, and full JSON response examples covering the common cases.	5 / 5
Workflow Clarity	Each command is clearly explained and "When to Use" is present, but the destructive `forget` is not presented as a verify-then-delete workflow despite a --dry-run flag existing, so the rubric's destructive-operation cap holds at 3.	3 / 5
Progressive Disclosure	Sections are well organized, but at ~288 lines everything (architecture diagram, config table, error handling) is inlined into SKILL.md with no one-level-deep reference files, and the bundled backend directories (mem0, memory_graph, memory_libsql) are not surfaced as navigable references.	3 / 5
	Total	14 / 20 Passed

Description

66%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.

The description conveys a clear purpose and a concrete command surface, with good trigger terms and distinctiveness, but it omits any explicit "Use when..." trigger guidance, which caps completeness.

Suggestions

Add a "Use when..." clause naming the situations that call for this skill (e.g., recalling past decisions, surfacing forgotten context mid-task) to lift completeness above 3.

Soften the metaphorical verbs ("absorbs", "surfaces") with one or two plain action verbs so the capability reads as concrete rather than poetic.

Include a natural user-side trigger phrase such as "Use when you need to remember or recall something across sessions."

Dimension	Reasoning	Score
Specificity	Lists several concrete actions ("Absorbs observations, surfaces associative memories as hypotheses, and builds context across sessions") plus five named commands; the verbs are slightly metaphorical, keeping it just below fully comprehensive.	4 / 5
Completeness	The "what" is clear (subconscious memory layer that absorbs/surfaces/builds), but there is no "Use when..." clause or equivalent trigger guidance, so completeness is capped at 3 per the rubric guideline.	3 / 5
Trigger Term Quality	Natural terms like "remember", "forget", "consult", "memory", and "context" appear, but it leans on command names rather than scenario phrasings a user would naturally say.	4 / 5
Distinctiveness Conflict Risk	The "subconscious/associative" framing plus the specific five-command set carves a fairly distinct niche, with only minor overlap risk against other generic memory tools.	4 / 5
	Total	15 / 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
metadata_field	'metadata' should map string keys to string values	Warning

	Total	15 / 16 Passed

Repository: Raudbjorn/cognitive-construct
Path: inland-empire/SKILL.md
Commit: 319f646

Reviewed: 6 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.