sublate

Vet working context by dialectical antithesis before action. Type: (ContextSuspect, User, VET, WorkingContext) → VettedContext

Quality

54%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./elenchus/skills/sublate/SKILL.md

Quality

Content

55%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 defines a rigorously sequenced protocol with explicit validation gates and feedback loops, but it is heavily verbose with repeated invariant restatements and inlines a large formal specification that would be better split into references.

Suggestions

Collapse the repeated invariant restatements across FLOW/MORPHISM/TYPES/PHASE TRANSITIONS/LOOP/MODE STATE into a single canonical statement, stating each point once and cross-referencing instead of re-arguing.

Move the formal type algebra (TYPES, MODE STATE, much of MORPHISM) into a references file (e.g., references/protocol-spec.md), leaving SKILL.md as an overview that links to it.

Front-load the actionable User-facing realization and Rules so a reader reaches concrete steps before parsing the abstract specification.

Dimension	Reasoning	Score
Conciseness	The body restates the same invariants (identity-vs-contents, the S_high conjunct, certificate-before-surfacing, J=dispositions) four to five times across FLOW, MORPHISM, TYPES, PHASE TRANSITIONS, LOOP, and MODE STATE, with long parenthetical justifications; this is noticeably verbose padding rather than the lean, assumes-competence anchor 5, though it is not teaching basic concepts Claude already knows.	2 / 5
Actionability	The User-facing realization gives a concrete presentation order, a literal Attribution-gate text template, a four-audit batch limit, and instruction consequences, but the bulk of the body is abstract type algebra and invariant prose rather than executable steps; this lands at 'some concrete guidance but incomplete', below the mostly-executable anchor 4.	3 / 5
Workflow Clarity	Phases are explicitly sequenced (Phase 0 -> 1 -> 2 -> 3, LOOP) with explicit validation checkpoints (fail-closed certificate, Qa attribution gate, Qs judgment gate) and a feedback loop (LOOP re-binds, re-certifies, re-posits on a met Revisit), matching the 'clear sequence with explicit validation and feedback loops' anchor.	5 / 5
Progressive Disclosure	One bundle file (references/round-composition.md) is well-signaled and linked at specific named moments, but the ~215-line FLOW/MORPHISM/TYPES/PHASES/LOOP/MODE STATE formal specification is inlined as a single monolithic block rather than split into a reference; this is 'some structure, content that should be separate is inline', below the well-split anchor 4.	3 / 5
	Total	13 / 20 Passed

Description

53%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 is concise and third-person with a clear 'what' and a distinctive niche, but it lacks an explicit 'when to use' trigger clause and leans on jargon ('dialectical antithesis') that users would rarely say naturally.

Suggestions

Add an explicit 'Use when ...' clause naming the concrete situations in which this skill should fire (e.g., before committing an action that depends on loaded context).

Replace or supplement 'dialectical antithesis' with natural trigger phrases a user would actually say, such as 'vet', 'check', or 'challenge my working context before acting'.

List 1-2 more concrete actions (e.g., 'challenges each claim's provenance, freshness, and inference') to lift specificity toward comprehensive coverage.

Dimension	Reasoning	Score
Specificity	Names the domain ("working context") and one concrete action ("Vet ... by dialectical antithesis before action"), but offers only a single action plus a type signature rather than comprehensive coverage; fits the '1-2 concrete actions' anchor and not the multi-action anchor 5.	3 / 5
Completeness	The 'what' is clear (vet working context by dialectical antithesis before action), but there is no 'Use when...' clause or equivalent explicit trigger guidance, which per the judging guidelines caps completeness at 3; it is not a 4 because the 'when' is entirely missing rather than weakly present.	3 / 5
Trigger Term Quality	Relevant keywords like "vet", "working context", and "before action" appear, but "dialectical antithesis" is philosophical jargon users would not naturally say, and common synonyms/variations are absent; sits at the 'some relevant keywords' anchor, below the strong natural-coverage anchor 5.	3 / 5
Distinctiveness Conflict Risk	The vetting-by-dialectical-antithesis framing plus the typed signature mark a clear niche with distinct triggers, but 'vetting context' still risks minor overlap with other review/validation skills, so it is 'mostly distinct' rather than the minimal-conflict anchor 5.	4 / 5
	Total	13 / 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: jongwony/epistemic-protocols
Path: elenchus/skills/sublate/SKILL.md
Commit: 3ffe94c

Reviewed: 1 day 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.