contextualize

Detect application-context mismatch after execution. Verifies applicability when correct output may not fit the actual context, registering each mismatch through a fail-closed deficit-fit certificate before disposition, producing contextualized execution. Judgment (does the aspect stand?) and disposition (adapt, keep, or withdraw the result) are separate axes: the judgment is relay-eligible on cited evidence, the disposition is the user's answer about the target in front of them, and a relay may close a mismatch only where the close leaves the result untouched. The transformative revalidation loop is non-monotone — an adapt disposition mutates the evaluation target and can breed emergent mismatches; re-scan is mandatory. Type: (ApplicationDecontextualized, AI, CONTEXTUALIZE, Result) → ContextualizedExecution. Alias: Epharmoge(ἐφαρμογή).

Quality

50%

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 ./epharmoge/skills/contextualize/SKILL.md

Quality

Content

41%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 content specifies a coherent, well-sequenced protocol with genuine validation and feedback loops and one good external reference, but it is crippled by extreme verbosity: a ~490-line inline formal block restates the pipeline redundantly and buries the actually executable surfacing templates and option wording. The formal machinery should be moved to a reference file and the body reduced to the actionable overview.

Suggestions

Move the formal blocks (TYPES, FLOW, PHASE TRANSITIONS, LOOP, CONVERGENCE, TOOL GROUNDING, MODE STATE, COMPOSITION) into a separate reference file (e.g. references/formal-spec.md) and keep SKILL.md as a lean overview that links to it, eliminating the redundant restatements of the same pipeline.

Lead the body with the actionable core — the Phase 0/1/2 walkthrough, the 'Done. One thing to verify:' surfacing template, the four option pairings, and the /gap /inquire /bound commands — and push the internal state-machine definitions behind a pointer.

Cut the exhaustive inline commentary that re-derives and re-justifies each invariant multiple times across sections; state each invariant once and reference it, trusting Claude to follow a stated rule without paragraph-length re-explanation.

Dimension	Reasoning	Score
Conciseness	The body is ~730 lines dominated by a ~490-line inline formal block that restates the same pipeline across FLOW, PHASE TRANSITIONS, LOOP, CONVERGENCE, TOOL GROUNDING, MODE STATE and COMPOSITION with exhaustive inline commentary — severely verbose and heavily padded with self-referential formalism far beyond what conveying the actionable behavior requires.	1 / 5
Actionability	There is concrete, copy-pasteable guidance — the 'Done. One thing to verify about applicability:' surfacing template, the four-option gate wording, the /gap /inquire /bound commands, and the severity table — but it is a minority of the file and buried under non-executable symbolic type machinery that defines internal state rather than instructing action.	3 / 5
Workflow Clarity	The protocol is clearly phased (Phase 0 silent scan, Phase 1 surfacing, Phase 2 disposition, LOOP re-scan) with explicit validation checkpoints (fail-closed certificate-before-registration, the ordered terminal check) and a real feedback loop (adapt → mandatory re-scan); the only friction is that the density makes the sequence hard to extract, not that checkpoints are missing.	4 / 5
Progressive Disclosure	Section headers give the document structure and references/round-composition.md is a well-signaled, one-level-deep reference linked from Rule 7 with clear occasion conditions, but the enormous formal-specification block (TYPES/FLOW/PHASE TRANSITIONS/LOOP/CONVERGENCE/TOOL GROUNDING/MODE STATE/COMPOSITION) is inlined wholesale in SKILL.md when it clearly belongs in a separate reference file.	3 / 5
	Total	11 / 20 Passed

Description

58%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 'what' and carves a distinct niche, but it is heavily padded with internal formal jargon, lacks natural user-facing trigger phrases, and omits any explicit 'Use when...' guidance. It reads as a protocol specification abstract rather than a trigger-oriented skill description.

Suggestions

Add an explicit 'Use when...' clause naming the natural situations a user would describe (e.g., 'Use when a technically correct result may not fit the actual project context, or when the user says the output doesn't fit their situation').

Replace or supplement internal jargon ('fail-closed deficit-fit certificate', 'transformative revalidation loop', 'ApplicationDecontextualized') with plain trigger terms users actually say ('context mismatch', 'doesn't fit', 'wrong assumptions', 'adapt or withdraw').

Lead with the concrete user-facing action (surface a context mismatch and offer to keep/adapt/withdraw the result) before the formal type signature and alias.

Dimension	Reasoning	Score
Specificity	Names several concrete protocol actions — 'Detect application-context mismatch', 'Verifies applicability', 'registering each mismatch through a fail-closed deficit-fit certificate', 'disposition (adapt, keep, or withdraw)', 're-scan is mandatory' — but the actions are abstract protocol operations wrapped in domain jargon rather than tangible operations, leaving minor gaps versus the fully concrete anchor 5.	4 / 5
Completeness	It clearly states the 'what' (detect/verify/register/dispose mismatches to produce contextualized execution), but there is no 'Use when...' clause or equivalent explicit trigger guidance — the 'after execution' cue only weakly implies when — so completeness is capped at 3 per the missing-trigger guideline.	3 / 5
Trigger Term Quality	Some relevant keywords appear ('application-context mismatch', 'may not fit the actual context', 'adapt, keep, or withdraw'), but the bulk is technical jargon ('fail-closed deficit-fit certificate', 'transformative revalidation loop', 'relay-eligible', 'ApplicationDecontextualized') and the natural phrases a user would actually say, plus common synonyms, are largely missing.	3 / 5
Distinctiveness Conflict Risk	The post-execution applicability/context-fit niche with its two-axis judgment/disposition model is mostly distinct from other skills, with only minor overlap risk against general result-review or verification skills; the highly specialized framing reduces false-trigger likelihood.	4 / 5
	Total	14 / 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
skill_md_line_count	SKILL.md is long (736 lines); consider splitting into references/ and linking	Warning

	Total	15 / 16 Passed

Repository: jongwony/epistemic-protocols
Path: epharmoge/skills/contextualize/SKILL.md
Commit: e56ccd7

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.