CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

pull-changes-resolve-conflicts

Standard workflow for pulling updates from main or other branches on multi-contributor projects (including Flows apps) without silently discarding work. Guides fetching/merging, requires listing merge conflicts explicitly, analyzing ours vs theirs using conversation history and repo context, presenting prioritized recommendations, and obtaining user answers before editing conflict markers or completing the merge. Triggers: pull main, merge main, merge origin, rebase, merge conflict, unmerged paths, both modified, integrate branch, sync with main, git merge abort, resolve conflicts, UU status, theirs vs ours, feat branch update.

78

1.00x
Quality

90%

Does it follow best practices?

Impact

65%

1.00x

1 of 3 eval scenarios. Add 2 more for a full score.

SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

85%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 strong, disciplined process skill: lean imperative guidance, concrete commands throughout, a well-sequenced workflow with explicit user-confirmation and post-resolution validation, and a useful anti-patterns section. Its one real defect is structural: it links to reference.md for optional command snippets and a merge message template, but that file does not exist in the bundle, leaving the only progressive-disclosure reference broken.

Suggestions

Create the referenced reference.md (or a references/ file) containing the optional command snippets and merge message template, or remove the dangling link — as written it points Claude to a file that does not exist.

Trim duplication between Goals and Hard rules (merge the 'do not assume main wins' points) and drop the repeated Flows/React app framing from either the intro or Quick reference to tighten conciseness toward 5.

In Quick reference, either cut the stage-2/stage-3 git-internal explanation (Claude already knows merge stage numbering) or reduce it to the single verification command.

DimensionReasoningScore

Conciseness

The body is policy/process content Claude does not already know (hard rules, prioritization scheme, abort handling) and it stays tight, with copy-paste commands rather than git tutorials. Minor trimmable spots keep it below 5: the Goals section partially restates Hard rules ("do not assume 'main wins'" vs "No silent resolution"), the intro's "Flows/React apps are a common case where conflicts cluster in app shells" is repeated in Quick reference, and the stage-2/stage-3 git semantics in Quick reference edge toward explanation of built-in git knowledge.

4 / 5

Actionability

For an instruction-only skill the guidance is fully concrete: exact commands ("git fetch" then "git merge origin/main", "git diff --name-only --diff-filter=U", "git show :2:path" vs "git show :3:path", "git merge --abort" / "git rebase --abort", "git checkout --conflict=merge <file>"), a concrete classification scheme (Orthogonal / Overlapping / Corruption risk), a P0–P3 ordering, and even a literal reply-format users can give ("(a) follow recommendations, (b) keep branch for file X..."). Per the rubric's code_vs_instruction note, absence of code is not penalized when guidance is this actionable.

5 / 5

Workflow Clarity

A clearly sequenced 5-step workflow with explicit checkpoints and a validation/feedback loop appropriate to this risky operation: pause at conflicts ("Stop at conflicts"), report before editing, analyze, "Only after the user confirms... apply resolutions", then validate ("Re-run git status; ensure no conflict markers remain; run tests or lint") with an explicit error path for aborting. The destructive-operation validation cap is satisfied, and the sequence matches the anchor-5 pattern of ordered steps plus explicit validation and recovery.

5 / 5

Progressive Disclosure

The body is well-sectioned (Goals, Hard rules, Workflow, Anti-patterns, Quick reference) and correctly pushes optional detail out of the main line ("For optional command snippets and a merge message template, see [reference.md](reference.md)"), but no references/, scripts/, or assets/ directories exist and reference.md is missing — the sole external reference is dangling, so the promised optional content is unreachable. This lands at anchor 3 ('references present but not clearly signaled / functional') rather than 4–5 where a real, loadable reference would put it.

3 / 5

Total

17

/

20

Passed

Description

96%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.

An excellent description: third-person voice, concrete multi-step capabilities, and an unusually thorough trigger list including git status codes and synonyms. The only weakness is that several broad triggers (rebase, sync with main) are not scoped to the conflict/integration situation, creating minor overlap risk with routine git-operation skills.

Suggestions

Scope the broadest triggers to the conflict context, e.g. 'rebase with conflicts', 'sync with main (conflicts)', so routine conflict-free rebases do not fire this skill.

Consider replacing the bare 'feat branch update' trigger with a more specific variant such as 'update feature branch from main' that a user would more naturally say.

DimensionReasoningScore

Specificity

Lists multiple concrete actions with comprehensive coverage: "Guides fetching/merging, requires listing merge conflicts explicitly, analyzing ours vs theirs using conversation history and repo context, presenting prioritized recommendations, and obtaining user answers before editing conflict markers or completing the merge." Anti-drift check: anchor 4 ('minor gaps in coverage') does not fit — no meaningful capability of the workflow is left unstated.

5 / 5

Completeness

Explicitly answers both: what ("Standard workflow for pulling updates... Guides fetching/merging, requires listing merge conflicts explicitly, analyzing ours vs theirs... presenting prioritized recommendations, and obtaining user answers before editing conflict markers") and when ("Triggers: pull main, merge main, ...resolve conflicts..."). The 'when' clause is explicit and uses concrete trigger phrases, matching the anchor-5 example structure exactly; anchor 4 ('when could be more explicit') is clearly below this.

5 / 5

Trigger Term Quality

The trigger list is comprehensive with natural phrasings, synonyms, and git status vocabulary: "pull main, merge main, merge origin, rebase, merge conflict, unmerged paths, both modified, integrate branch, sync with main, git merge abort, resolve conflicts, UU status, theirs vs ours, feat branch update" — these are exactly what a user or transcript would contain during a conflicted pull. Not 4: no common variation of the request is conspicuously missing.

5 / 5

Distinctiveness Conflict Risk

The niche (conflict-aware integration of another branch) is clear and most triggers ('merge conflict', 'UU status', 'unmerged paths', 'theirs vs ours') are distinct to it. However, broad triggers like "rebase", "sync with main", and "feat branch update" would also fire on routine, conflict-free rebases or branch updates handled by ordinary git skills — minor overlap risk, matching anchor 4 rather than anchor 5's 'minimal conflict risk'.

4 / 5

Total

19

/

20

Passed

Validation

81%

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

Validation — 13 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

allowed_tools_field

'allowed-tools' contains unusual tool name(s)

Warning

metadata_version

'metadata.version' is missing

Warning

relative_links

Relative link issues: 1 missing

Warning

Total

13

/

16

Passed

Repository

cognitedata/builder-skills

Path

skills/pull-changes-resolve-conflicts/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.