CtrlK
BlogDocsLog inGet started
Tessl Logo

gh-stack

Manages stacked PRs and splits multi-part work into reviewable branches with gh-stack. Use for stack creation, viewing, edits, push, submit, sync, rebase, merge, or checkout; when asked to split or isolate work for review; whenever a user mentions a stack, branch layers, dependent PRs, or gh stack; or when a stack is checked out.

77

Quality

96%

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

The canonical home for this skill is gh-stack in Arize-ai/phoenix

SKILL.md
Quality
Evals
Security

Quality

Content

92%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 an excellent operational reference: fully executable commands, non-interactive pitfalls, exit codes with recovery, and well-signaled progressive disclosure to real reference files. Its only weakness is minor trimmable detail (merge-queue semantics, JSON schema, repeated --remote warning) that could move to references/commands.md.

DimensionReasoningScore

Conciseness

The body is dense and operational with no padding or explanations of known concepts, but a few details could be trimmed or moved to references: the merge-queue paragraph, the JSON schema block, and the --remote caveat which is stated twice. Anchor 4 (minor instances that could be trimmed) fits; it is not noticeably verbose (3) nor perfectly lean (5).

4 / 5

Actionability

Fully executable guidance throughout: copy-paste bash blocks for the core loop, branch placement, sync, and conflict recovery; an always-run/never-run table with the reason each bare command fails; and a per-exit-code recovery column. Common cases including rebase conflicts are covered.

5 / 5

Workflow Clarity

Sequences are explicit (checkout owner → edit → commit → rebase upstack → top → push) with a validation step ("gh stack view --json # confirm") and full feedback loops: the exit-code table pairs every failure with a recovery command, and Exit 3 has continue/abort paths for both rebase and sync. The destructive/batch cap does not apply since validation is present.

5 / 5

Progressive Disclosure

A lean overview with three verified, one-level-deep references, each carrying an explicit read-when trigger ("read before creating a stack", "read when a command fails unexpectedly", "read on a rebase conflict") plus the guidance to open only the matching one. This trigger-based navigation matches or exceeds the anchor-5 example.

5 / 5

Total

19

/

20

Passed

Description

100%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 exemplary: third-person, concrete, comprehensive over the tool's operations, with explicit and multi-form trigger guidance including a checked-out-state trigger. It closely matches the rubric's good_overall_examples pattern.

DimensionReasoningScore

Specificity

"Manages stacked PRs and splits multi-part work into reviewable branches" plus the enumerated verb list ("stack creation, viewing, edits, push, submit, sync, rebase, merge, or checkout") gives comprehensive concrete-action coverage of the tool, matching the anchor-5 example structure.

5 / 5

Completeness

Explicitly answers both what ("Manages stacked PRs and splits multi-part work into reviewable branches") and when ("Use for...; whenever a user mentions a stack...; or when a stack is checked out") with concrete trigger phrases, mirroring the anchor-5 example.

5 / 5

Trigger Term Quality

Natural terms users would say are comprehensively covered: "stack", "stacked PRs", "gh stack", "branch layers", "dependent PRs", and "split or isolate work for review"; no file extensions exist for this domain, so synonym coverage is complete.

5 / 5

Distinctiveness Conflict Risk

The gh-stack tool name and stack-specific nouns dominate the triggers, including a state-based trigger ("when a stack is checked out"), giving a clear niche with minimal conflict risk; the broadest clause ("split or isolate work for review") still accurately describes this tool's purpose.

5 / 5

Total

20

/

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
trackdubllc/Trackdub
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.