fla-design-coverage

Contract-first design and coverage discipline for FLA kernel and numerical changes. Use before implementation to define supported cells, numerical budgets, dispatch semantics, compatibility, tests, and benchmarks.

Quality

74%

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 ./.agents/skills/fla-design-coverage/SKILL.md

Quality

Content

63%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 is highly actionable and well-sequenced with concrete commands and validation hooks, but it is a dense single-file monolith that could benefit from tighter prose and offloading detail into reference files.

Suggestions

Tighten repeated invariant statements (e.g. the Existing-fallback and tolerance-change rules appear across sections) to reduce length and improve conciseness.

Move the numerical-budget field table and coverage-layer table into a references/ file linked from SKILL.md so the body stays a lean overview with one-level-deep navigation.

Make validation checkpoints explicit feedback loops (e.g. 'run verify; if it fails, fix and re-run before landing the candidate') rather than implicit conditions.

Dimension	Reasoning	Score
Conciseness	The body avoids explaining basics Claude already knows, but is a long, dense monolithic block of prescriptive contract prose that could be tightened; several passages restate the same invariants.	3 / 5
Actionability	Gives concrete, executable guidance — specific commands ('python -m benchmarks.ops.verify --op <op>'), file paths ('tests/ops/test_<op>.py'), env flags ('FLA_TILELANG=0/1'), and a real formula pattern — with only minor abstract gaps.	4 / 5
Workflow Clarity	Opens with an ordered 5-step pre-implementation sequence and includes validation checkpoints (verify stays green, parity tests, route-parity), though some checkpoints are implicit rather than spelled out as explicit feedback loops.	4 / 5
Progressive Disclosure	No bundle files exist (references/scripts/assets absent) and all detail — including the budget and coverage-layer tables — is inlined in one file; cross-skill references like fla-correctness-coverage are named but not organized as one-level-deep navigation.	3 / 5
	Total	14 / 20 Passed

Description

85%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 specific, complete, and clearly niche-scoped, with an explicit 'Use before implementation' trigger. Its main weakness is trigger-term breadth: it relies on project jargon without synonyms a user might naturally say.

Dimension	Reasoning	Score
Specificity	Names the domain and enumerates multiple concrete actions — 'define supported cells, numerical budgets, dispatch semantics, compatibility, tests, and benchmarks' — giving comprehensive coverage rather than vague language.	5 / 5
Completeness	Explicitly answers both what ('Contract-first design and coverage discipline for FLA kernel and numerical changes') and when ('Use before implementation to define ...'), with a concrete trigger phrase.	5 / 5
Trigger Term Quality	Includes relevant domain keywords ('FLA kernel', 'numerical changes', 'before implementation') but lacks common synonyms or variations a user might naturally say, leaning on project jargon.	3 / 5
Distinctiveness Conflict Risk	Targets a clear niche — FLA kernel and numerical contract design — with triggers unlikely to fire for unrelated skills, so conflict risk is minimal.	5 / 5
	Total	18 / 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: fla-org/flash-linear-attention
Path: .agents/skills/fla-design-coverage/SKILL.md
Commit: f470469

Reviewed: 4 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.