conformance-loop

Iteratively improve Fallow analysis accuracy by comparing it with competing tools and verified source truth across a stable real-world corpus.

Quality

66%

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 ./.claude/skills/conformance-loop/SKILL.md

Quality

Content

78%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 tight, well-sequenced conformance workflow that is maximally concise and appropriately structured for a short single-purpose skill, with its main weakness being high-level steps that lack the concrete execution details needed for full actionability.

Suggestions

Make 'documented equivalent settings' concrete by naming the settings to match (e.g., sensitivity thresholds, rule sets) or pointing to where they are recorded.

Specify the regression fixture format and where it lives so step 5 is executable rather than aspirational.

Add an explicit re-run-on-regression failure branch (e.g., 'If the correction regresses other cases, discard it and re-run') to turn step 6's implicit gate into a clear feedback loop.

Dimension	Reasoning	Score
Conciseness	The body is a lean seven-step list with no padding and no explanation of concepts Claude already knows; the closing maxim ('Competitor output is a lead, not ground truth.') earns its place, matching the 'lean and efficient' anchor.	5 / 5
Actionability	It offers some concrete guidance (the TP/FP/FN/model-difference labels and the literal `review` command) but several steps stay high-level ('documented equivalent settings', 'a regression fixture') without the specifics to execute, matching the 'some concrete guidance but incomplete' anchor.	3 / 5
Workflow Clarity	Steps 1–7 are clearly sequenced with checkpoints (step 3 verifies disagreements, step 6 retains only net improvements as a gate) for this batch/corpus operation; not a 5 because error-recovery feedback on a failed re-run is only implicit.	4 / 5
Progressive Disclosure	Under 50 lines, single-purpose, no external references needed, and organized as a clean numbered workflow plus a one-line principle, so the simple-skill exception applies and progressive disclosure is fully appropriate.	5 / 5
	Total	17 / 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 clearly states what the skill does in a specific niche, but omits any 'when to use' trigger guidance, which caps completeness and leaves trigger terms domain-jargon-heavy rather than user-natural.

Suggestions

Add a 'Use when...' clause naming concrete trigger phrases (e.g., 'Use when improving Fallow detection accuracy, validating Fallow against other tools, or auditing a corpus of disagreements').

Include natural-language synonyms a user might actually say (e.g., 'Fallow results', 'false positives', 'comparison run') alongside the technical jargon.

Add a brief mention of the input/output (a project corpus and a retained set of net improvements) so the 'what' is concrete rather than purely methodological.

Dimension	Reasoning	Score
Specificity	Names the domain ('Fallow analysis accuracy') and a couple concrete actions ('comparing it with competing tools and verified source truth across a stable real-world corpus'), but coverage is not comprehensive enough for a 4.	3 / 5
Completeness	Gives a clear 'what' (iteratively improve accuracy via comparison) but provides no 'Use when...' trigger clause, so per the judging guidelines completeness is capped at 3.	3 / 5
Trigger Term Quality	Contains relevant domain keywords ('Fallow analysis', 'competing tools', 'source truth', 'corpus') but leans on technical jargon and lacks the natural synonyms/variations a user would say, fitting the 'some relevant keywords but missing common variations' anchor.	3 / 5
Distinctiveness Conflict Risk	The Fallow-analysis conformance niche is mostly distinct with only minor overlap risk against general analysis skills; not a 5 because no explicit trigger phrases lock it to its niche.	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: fallow-rs/fallow
Path: .claude/skills/conformance-loop/SKILL.md
Commit: 7e1e075

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.