CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

new-target-backend

Add a new target backend skeleton to LLVM 22. Covers CMake registration, TargetInfo, TargetMachine, MCTargetDesc, minimal TableGen register/instruction defs, and wiring into the LLVM build.

Quality

78%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./tiles/tessl-llvm/skills/new-target/SKILL.md
SKILL.md
Quality
Evals
Security

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.

The body is highly actionable, well-sequenced, and token-efficient, but it is monolithic — everything is inlined in SKILL.md with no one-level-deep reference files to split out the per-layer templates.

Suggestions

Move the large per-file C++/CMake/TableGen templates into reference files (e.g. references/MCTargetDesc.cpp, references/TargetMachine.cpp) and keep SKILL.md as an overview that links to them.

Add an explicit error-recovery loop in Step 8 (e.g. 'If registration check fails, re-verify Triple.h/Triple.cpp edits and the LLVMInitialize* calls, then rebuild').

Trim redundant inline comments such as the data-layout note where the guidance is already implied by 'adjust for your ISA'.

DimensionReasoningScore

Conciseness

Mostly lean code with brief step prose and little concept over-explanation, though a few inline comments and asides could be trimmed.

4 / 5

Actionability

Provides copy-paste-ready CMake, TableGen, C++, and build/verify commands covering the common skeleton case end to end.

5 / 5

Workflow Clarity

Steps 0–8 are clearly sequenced with explicit verification in Step 8 and a 'Common mistakes' checklist, but there is no explicit fix→re-validate feedback loop.

4 / 5

Progressive Disclosure

No bundle files exist and all per-file templates are inlined in a single ~400-line SKILL.md despite good section structure; content that could live in separate reference files is inline.

3 / 5

Total

16

/

20

Passed

Description

67%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 and distinctive, naming concrete LLVM backend layers, but omits an explicit 'Use when...' trigger and leans on technical jargon over natural user phrasing.

Suggestions

Add an explicit 'Use when...' clause, e.g. 'Use when the user wants to add a new ISA target or backend to LLVM.'

Include natural synonyms a user might say — 'backend', 'ISA', 'target architecture' — alongside the layer names.

Keep the layer list but consider leading with the user-facing goal rather than the internal component names.

DimensionReasoningScore

Specificity

Lists six concrete actions — CMake registration, TargetInfo, TargetMachine, MCTargetDesc, minimal TableGen register/instruction defs, and build wiring — giving comprehensive coverage of the skeleton layers.

5 / 5

Completeness

Clearly states what the skill does, but lacks an explicit 'Use when...' trigger clause; per the judging guidelines a missing trigger clause caps completeness at 3.

3 / 5

Trigger Term Quality

Contains one natural phrase ('add a new target backend') but the remaining terms (MCTargetDesc, TargetInfo, TableGen) are LLVM jargon users rarely say, and synonyms like 'backend' or 'ISA' are absent.

3 / 5

Distinctiveness Conflict Risk

Targets a very specific niche (adding a new ISA target backend to LLVM 22) with distinct triggers and minimal overlap risk with other skills.

5 / 5

Total

16

/

20

Passed

Validation

100%

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

hbarve1/tessl-llvm

Path

tiles/tessl-llvm/skills/new-target/SKILL.md

Revision

a251bde

Reviewed

Table of Contents

DiscoveryImplementationValidation

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.

Claim skill