CtrlK
BlogDocsLog inGet started
Tessl Logo

pr-slicer

This skill should be used when the user asks to "split this PR", "make this smaller", "create stacked PRs", "slice this change", or "reduce review churn". Helps break work into low-risk, reviewable slices with clear verification per slice.

70

Quality

88%

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

SKILL.md
Quality
Evals
Security

Quality

Content

86%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-structured instruction-only skill: concrete budgets, an explicit slicing order, and a disciplined handoff format with no wasted tokens. The gaps are illustrative — one worked example slice packet and one concrete verification command would make the guidance fully copy-paste ready, and an explicit failure/retry path would complete the workflow.

Suggestions

Add one filled-in example slice packet (goal, owned files, out of scope, risk level, an actual verification command with its expected pass condition, rollback plan) to make the template concrete.

Specify what "project gate checks" means or give an example (e.g. 'the repo's CI test/lint target') so verification is executable rather than inferred.

Add an explicit failure path, e.g. 'If a slice exceeds budget or verification fails, re-split it before proceeding', to close the feedback loop.

DimensionReasoningScore

Conciseness

The body is lean and entirely prescriptive — budgets, ordered steps, template fields, rules, and output format with zero padding and no explanation of concepts Claude already knows (what a PR is, why small reviews help beyond one goal line). Level 4 would require minor over-explanation to trim, which is absent.

5 / 5

Actionability

Concrete, executable guidance dominates: specific budgets ("<= 400 changed lines", "<= 10 changed files", "1-4 commits"), a slicing order, and a packet template with defined fields. It falls short of level 5 because verification guidance stays generic ("Run project gate checks" with no example command) and no filled-in example slice packet is provided.

4 / 5

Workflow Clarity

The Slicing Order gives a clear 4-step sequence, and Verification Discipline plus the packet's risk level and rollback plan supply most checkpoints. It does not reach level 5 because there is no explicit feedback loop (what to do when a slice's verification fails or a budget is exceeded) and no complete checklist.

4 / 5

Progressive Disclosure

The body is under 50 lines with no external references needed and clean, well-labeled sections in a logical order (goal, budgets, order, template, rules, verification, handoff). Per the simple-skill guideline this earns the top anchor; the body is short enough that no file split is warranted.

5 / 5

Total

18

/

20

Passed

Description

86%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.

A strong description with excellent explicit trigger phrases and a clear what/when structure in third-person voice. Its only weakness is that the capability description itself is thin — it summarizes the outcome (slices with verification) without enumerating the concrete actions the skill performs.

Suggestions

Enumerate 1-2 more concrete actions in the 'what' clause, e.g. 'Orders slices by dependency (prerequisites, refactors, behavior changes), drafts a per-slice packet with owned files, risk level, verification commands, and rollback plan'.

Add one or two more domain-specific trigger variants such as 'stacked diffs' or 'break up this PR' to reduce residual overlap with generic refactoring requests.

DimensionReasoningScore

Specificity

The description names the domain and about two concrete actions — "break work into low-risk, reviewable slices" and "clear verification per slice" — but does not comprehensively cover what the skill does (no mention of slicing order, budgets, or handoff output). It is more than the minimal/generic level 2 but lacks the several distinct actions of level 4.

3 / 5

Completeness

It explicitly answers both parts: "should be used when the user asks to..." (when, with concrete trigger phrases) and "Helps break work into low-risk, reviewable slices with clear verification per slice" (what). This matches the anchor for a clear and explicit what AND when with concrete triggers; level 4 would require the 'when' to be less explicit.

5 / 5

Trigger Term Quality

It quotes five natural user phrasings with synonyms — "split this PR", "make this smaller", "create stacked PRs", "slice this change", "reduce review churn" — comprehensive natural trigger coverage that a user would plausibly say verbatim. Level 4 would apply if common variants were missing, but the synonym coverage is broad for this domain.

5 / 5

Distinctiveness Conflict Risk

The niche (stacked PRs / change slicing) is mostly distinct with dedicated triggers, though broader phrases like "make this smaller" and "reduce review churn" carry minor overlap risk with general refactoring or code-review skills. Not level 5 because not all triggers are unique to this skill.

4 / 5

Total

17

/

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
spacedriveapp/spacebot
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.