CtrlK
BlogDocsLog inGet started
Tessl Logo

lightrun-runtime-aware-pr-review

Use when reviewing a pull request with runtime or production evidence — for example to review a PR with runtime verification, gather production evidence, or simulate a patch on live samples. Reviews a pull request by diffing against the PR merge base, collecting live samples, and simulating the patch on captured production inputs.

90

1.38x
Quality

96%

Does it follow best practices?

Impact

98%

1.38x

Average score across 2 eval scenarios

SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

92%Weight 40%Scale 1-3

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

A well-structured multi-phase process skill with exemplary sequencing, validation checkpoints, and feedback loops; the only real weakness is bundle hygiene — orphaned reference files with inconsistent phase numbering that could mislead navigation.

Suggestions

Delete the orphaned reference files not linked from the body (phase-2-runtime-profile.md, phase-3-sampling-plan.md, phase-4-execute-snapshots.md, phase-5-patch-simulation.md) so the bundle matches the body's phase numbering and navigation.

Reconcile the two parallel phase-numbering schemes in references/ — the body uses Phase 3=runtime-profile through Phase 6=patch-simulation, while the orphaned files use Phase 2–5 — to prevent an agent browsing the directory from following stale instructions.

DimensionReasoningScore

Conciseness

Lean overview that assumes Claude's competence (no explanations of PRs, diffs, or SHAs) and uses domain terms directly; the Checkpoints restate per-phase exit criteria but as a tight consolidated gate list rather than prose.

3 / 3

Actionability

Concrete guidance throughout — named artifacts ('pr_base_sha', 'pr_head_sha'), explicit rules ('Never use deployed_sha → pr_head_sha as the PR review diff'), and numeric caps ('up to 3') — with executable detail appropriately delegated to phase references.

3 / 3

Workflow Clarity

Phases 0–6 are explicitly sequenced with a dedicated Checkpoints section, feedback loops ('loop back to the prior phase when evidence is incomplete'), and error-recovery (emit Sampling Request and stop on zero hits).

3 / 3

Progressive Disclosure

The body is a clear overview with well-signaled one-level-deep links (all 10 resolve), but references/ holds 4 orphaned files (phase-2-runtime-profile.md, phase-3-sampling-plan.md, phase-4-execute-snapshots.md, phase-5-patch-simulation.md) using a stale numbering scheme, so the bundle could be better organized.

2 / 3

Total

11

/

12

Passed

Description

100%Weight 40%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.

A strong, specific description with an explicit 'Use when' trigger covering both what the skill does and when to use it; third-person voice is correct and the trigger terms are natural for the target audience.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — 'diffing against the PR merge base, collecting live samples, and simulating the patch on captured production inputs' — matching the multiple-specific-actions anchor.

3 / 3

Completeness

Explicitly answers both what ('Reviews a pull request by diffing...') and when ('Use when reviewing a pull request with runtime or production evidence'), satisfying the what-and-when anchor.

3 / 3

Trigger Term Quality

Natural user-facing terms include 'reviewing a pull request,' 'runtime,' 'production evidence,' and 'live samples,' giving good coverage of phrases a user would actually say.

3 / 3

Distinctiveness Conflict Risk

The runtime/production-evidence niche and triggers like 'runtime verification' and 'simulate a patch on live samples' are clearly distinct from a generic PR-review skill.

3 / 3

Total

12

/

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.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
lightrun-platform/lightrun-ai
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.