CtrlK
BlogDocsLog inGet started
Tessl Logo

add-gc-statepoints

Add garbage-collection support to an LLVM 22 IR frontend. Covers the shadow-stack (gcroot) model for simple GCs and the statepoint/gc.relocate model for moving/relocating collectors, including the RewriteStatepointsForGC pass and StackMap section.

67

Quality

80%

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/add-gc-statepoints/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

92%Scale 1-3

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

This is a high-quality skill that efficiently covers two distinct GC integration models with concrete, executable code and clear step-by-step workflows. The decision table at the top provides immediate guidance, and the 'Common mistakes' section adds valuable guardrails. The only minor weakness is that all content is inline rather than using progressive disclosure for advanced topics, though the overall length is manageable.

DimensionReasoningScore

Conciseness

The skill is lean and efficient throughout. It assumes Claude knows LLVM, C++, and IR concepts without explaining them. Every section delivers actionable content without padding. Comments in code are minimal and purposeful.

3 / 3

Actionability

Provides fully executable C++ and LLVM IR code snippets that are copy-paste ready. Includes specific API calls (getOrInsertDeclaration, CreateAlloca, addPass), concrete struct definitions for the runtime, and exact CLI commands for verification.

3 / 3

Workflow Clarity

Both models are presented as clearly numbered step sequences (A1-A4, B1-B5). The statepoint model includes the RewriteStatepointsForGC pass workflow with a clear 4-step description of what it does. Verification steps are explicit with concrete commands, and the 'Common mistakes' section serves as a validation checklist for error prevention.

3 / 3

Progressive Disclosure

The content is well-structured with clear sections and a decision table upfront, but it's entirely self-contained in one file. The manual statepoint emission (Step B3) and custom GC strategy (Step B5) could be split into separate reference files to keep the main skill leaner. However, for a skill of this length (~120 lines), inline content is reasonable.

2 / 3

Total

11

/

12

Passed

Description

67%Scale 1-3

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 highly specific and technically detailed, clearly carving out a distinct niche around LLVM IR garbage-collection support. Its main weakness is the absence of an explicit 'Use when...' clause, which would help Claude know exactly when to select this skill. The trigger terms are accurate but lean heavily on technical jargon without covering more natural user phrasings.

Suggestions

Add an explicit 'Use when...' clause, e.g., 'Use when the user asks about adding garbage collection to an LLVM frontend, implementing gcroot/shadow-stack, statepoint-based GC, or the RewriteStatepointsForGC pass.'

Include more natural trigger term variations such as 'GC support', 'LLVM garbage collector', 'managed memory', or 'GC safepoints' to improve matching against diverse user queries.

DimensionReasoningScore

Specificity

Lists multiple specific concrete actions and concepts: shadow-stack (gcroot) model, statepoint/gc.relocate model, RewriteStatepointsForGC pass, and StackMap section. These are precise, named capabilities.

3 / 3

Completeness

Clearly answers 'what does this do' (add GC support to LLVM IR frontend covering two models and related passes), but lacks an explicit 'Use when...' clause or equivalent trigger guidance, which caps this at 2 per the rubric.

2 / 3

Trigger Term Quality

Includes strong domain-specific terms like 'garbage-collection', 'LLVM', 'gcroot', 'statepoint', 'gc.relocate', 'shadow-stack', and 'StackMap'. However, these are highly technical and may miss more natural user phrasings like 'GC support', 'IR frontend GC', or 'LLVM garbage collector integration'. Coverage of natural variations is limited.

2 / 3

Distinctiveness Conflict Risk

Extremely specific niche: LLVM 22 IR frontend garbage-collection support. This is unlikely to conflict with any other skill due to the highly specialized domain and explicit mention of LLVM version, GC models, and specific passes.

3 / 3

Total

10

/

12

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.

Validation11 / 11 Passed

Validation for skill structure

No warnings or errors.

Repository
hbarve1/tessl-llvm
Reviewed

Table of Contents

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.