CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

gh-stack

Manage stacked branches and pull requests with the gh-stack GitHub CLI extension. Use when the user wants to create, push, rebase, sync, navigate, or view stacks of dependent PRs. Triggers on tasks involving stacked diffs, dependent pull requests, branch chains, or incremental code review workflows.

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 ./.agents/skills/gh-stack/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

65%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 exceptionally actionable command reference with concrete flags, error-recovery loops, and output schemas — top marks on actionability. It loses points for token-inefficient filler code and triple-stated rules, an unverified destructive unstack workflow, and a monolithic single-file layout that inlines a full command reference instead of splitting it into bundle files.

Suggestions

Remove the placeholder heredoc Go source files from the workflows and consolidate the non-interactivity rules (currently stated in Agent rules, the 'Never do' list, and per-command sections) into one authoritative location to cut substantial tokens.

Add a verification checkpoint to the destructive 'unstack' restructure workflow — e.g., run 'gh stack view --json' to confirm PR merge state before tearing down the stack locally and on GitHub.

Move the detailed per-command reference and the --json output schema into a references/ file (e.g., COMMANDS.md), keeping the quick-reference table and workflows in SKILL.md with clearly signaled one-level-deep links.

DimensionReasoningScore

Conciseness

The bulk is dense, tool-specific reference that Claude could not know, but ~80 lines of filler heredoc Go source files in the workflows add nothing about gh-stack, and the non-interactivity rules are repeated three times (Agent rules, the 'Never do' list, and per-command notes). Mostly efficient with clearly trimmable sections — matches anchor 3 better than the noticeably-padded anchor 2.

3 / 5

Actionability

Fully executable, copy-paste-ready guidance throughout: exact commands with flags, a quick-reference table, the full --json output schema, jq parsing recipes, exit-code-to-agent-action mapping, and specific conflict-resolution steps. Covers the common cases comprehensively.

5 / 5

Workflow Clarity

Sequencing is excellent and the rebase-conflict workflow has a full validate/fix/retry feedback loop with --continue/--abort, but the destructive 'unstack' restructure workflow (tears down the stack locally and on GitHub) includes no verification checkpoint such as confirming PR merge state via 'gh stack view --json' first. Per the rubric, a destructive workflow without validation caps workflow clarity at 3 — this takes precedence over the otherwise strong sequencing.

3 / 5

Progressive Disclosure

No bundle files exist (references/, scripts/, assets/ are absent), so everything lives in a single ~780-line SKILL.md. Internal structure is good (headers, quick-reference table), but ~400 lines of full per-command reference plus the JSON schema are inlined content that would be better split into a referenced file. Some structure with content that should be separate inline — anchor 3; not 2 because navigation via the table and headers is genuinely easy.

3 / 5

Total

14

/

20

Passed

Description

83%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 a clear what/when structure, concrete action verbs, and mostly natural trigger terms. It is held back from top marks only by minor coverage gaps (submit/teardown actions) and a slightly over-broad 'incremental code review workflows' trigger phrase.

DimensionReasoningScore

Specificity

Names the tool ('gh-stack GitHub CLI extension') and lists six concrete actions ('create, push, rebase, sync, navigate, or view stacks of dependent PRs'), but omits PR creation via submit and stack teardown/restructure — minor gaps in coverage that keep it below the comprehensive anchor.

4 / 5

Completeness

Clearly answers both: 'what' ('Manage stacked branches and pull requests with the gh-stack GitHub CLI extension') and explicit 'when' with concrete trigger phrases ('Use when the user wants to... Triggers on tasks involving stacked diffs, dependent pull requests, branch chains'). Matches the anchor-5 example structure directly.

5 / 5

Trigger Term Quality

Good natural-term coverage ('stacked diffs', 'dependent pull requests', 'branch chains', 'stacks of dependent PRs'), but common variants users would actually say such as 'stacked PRs' or 'PR stack' are missing. Not score 3 because multiple relevant natural keywords are present; not 5 because synonym coverage is incomplete.

4 / 5

Distinctiveness Conflict Risk

A clear niche (stacked branch/PR management) with mostly distinct triggers, but 'incremental code review workflows' is a broad phrase that could overlap with general code-review skills. Not 5 due to that minor overlap risk; well above 3 since the core triggers are unmistakably about PR stacks.

4 / 5

Total

17

/

20

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

skill_md_line_count

SKILL.md is long (795 lines); consider splitting into references/ and linking

Warning

Total

15

/

16

Passed

Repository

phatblat/dotfiles

Path

.agents/skills/gh-stack/SKILL.md

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.