grasp

Verify understanding after AI work through intent-scented entry points. Type: (ResultUngrasped, User, VERIFY, Result) → VerifiedUnderstanding. Alias: Katalepsis(κατάληψις).

Quality

48%

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 ./katalepsis/skills/grasp/SKILL.md

Quality

Content

47%

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

The body defines a clearly sequenced, well-gated verification workflow but is a monolithic, heavily formalized ~500-line document that pays a large token cost for apparatus an intelligent reader does not need. Workflow clarity is strong; conciseness is the clear weakness.

Suggestions

Move the formal block (FLOW/MORPHISM/TYPES/PHASE TRANSITIONS/COMPOSITION) and the taxonomies into separate reference files (e.g. references/formal-spec.md, references/taxonomies.md), leaving SKILL.md as a concise overview with one-level-deep links.

Cut the repeated restatements of the Horizon/admissible(HC) logic so it is defined once and referenced thereafter; remove Greek-rooted narrative terms in favor of plain English.

Replace the abstract type-signature phase transitions with a short numbered plain-English walkthrough, keeping the concrete TaskCreate and question templates inline.

Dimension	Reasoning	Score
Conciseness	The ~500-line body is densely padded with formal notation (morphisms, type signatures, Greek-rooted terms) and restates the Horizon logic across the formal block, phase transitions, gap taxonomy, and rules, matching the score-1 'verbose; padded with unnecessary context' anchor.	1 / 3
Actionability	Concrete templates exist (TaskCreate blocks, question/options YAML, Entry Point and Gap taxonomies) but the bulk is abstract formal specification rather than copy-paste-ready instructions, fitting the score-2 'some concrete guidance but incomplete' anchor.	2 / 3
Workflow Clarity	Phases 0–3 are explicitly sequenced with numbered steps, validation gates (zero-gap confirmation, coverage check) and feedback loops (misconception 3-step re-verification), matching the score-3 'clear sequence with explicit validation steps; feedback loops for error recovery' anchor.	3 / 3
Progressive Disclosure	No bundle files exist; the skill is a single monolithic ~500-line file with clear section headers but content (the formal block, taxonomies) that should be split into reference files remains inline, fitting the score-2 'some structure but content that should be separate is inline' anchor.	2 / 3
	Total	8 / 12 Passed

Description

50%

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 distinct comprehension-verification purpose but leans on formal type-theoretic notation and a Greek alias rather than natural user language, and it omits an explicit 'Use when' trigger clause. All four dimensions land at the middle anchor as a result.

Suggestions

Add an explicit trigger clause, e.g. 'Use when the user says "help me understand", "what did you change?", "walk me through it", or wants to verify their grasp of recent AI work.'

Drop or relocate the formal type signature '(ResultUngrasped, User, VERIFY, Result) → VerifiedUnderstanding' and the 'Alias: Katalepsis(κατάληψις)' from the description so the frontmatter stays in natural user vocabulary.

Replace 'intent-scented entry points' with plain phrasing like 'offering the user a choice of what to verify first' to make the action concrete.

Dimension	Reasoning	Score
Specificity	Names a concrete action and domain ("Verify understanding after AI work through intent-scented entry points") but 'intent-scented entry points' is jargon and only a single action is named, not multiple concrete actions, matching the score-2 anchor.	2 / 3
Completeness	It clearly answers what the skill does but offers no explicit 'Use when...' trigger guidance; 'after AI work' only implies the when, so completeness caps at 2 per the judging guidelines.	2 / 3
Trigger Term Quality	A few natural terms appear ('understanding', 'AI work', 'verify') but the description is dominated by the formal type signature '(ResultUngrasped, User, VERIFY, Result) → VerifiedUnderstanding' and the Greek alias 'Katalepsis(κατάληψις)', missing common variations like 'help me understand' or 'explain this'.	2 / 3
Distinctiveness Conflict Risk	The comprehension-verification niche is distinct from explanation skills, but weak natural triggers and heavy jargon leave it overlapping with general 'explain/understand' skills, fitting the score-2 'somewhat specific but could overlap' anchor.	2 / 3
	Total	8 / 12 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
skill_md_line_count	SKILL.md is long (504 lines); consider splitting into references/ and linking	Warning

	Total	15 / 16 Passed

Repository: jongwony/epistemic-protocols
Path: katalepsis/skills/grasp/SKILL.md
Commit: ab9aa56

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.