major-task

Work heavyweight framework or library tasks with planning-first research, selective deep analysis, and rigorous handoff

Quality

60%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/major-task/SKILL.md

Quality

Content

71%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 dense, disciplined orchestration skill: concrete commands, ordered per-lane workflows, explicit verification gates, and clear conditional boundaries, with almost no token waste in the prose itself. Its two real weaknesses are structural — a 345-line single-file body that inlines lane-specific and tracker-specific material which belongs in reference files, and duplicated handoff/sync contract text repeated across four sections.

Suggestions

Move lane-specific detail into reference files (e.g. references/tracker-rules.md for GitHub/Linear sync-back and changeset blocks, references/editor-candidates.md for the Plate/Slate/Lexical mapping) and keep SKILL.md as the overview, which would cut the body roughly in half.

State the PR/tracker-sync and final-handoff contract once in a single section and reference it from the other sections instead of restating it in Tracked Task Rules, Verification, Post Back To Tracker, and Final Handoff.

Add an explicit error-recovery loop for code-changing work (e.g. "if targeted tests fail, fix and re-run before creating the PR") to close the workflow-clarity gap.

Dimension	Reasoning	Score
Conciseness	The body is terse, imperative, and never explains concepts Claude already knows, but the PR/tracker-sync contract is restated in Tracked Task Rules, Verification ("If verified work changed code, create or update the PR before tracker sync-back"), Post Back To Tracker, and Final Handoff, and editor-candidate guidance repeats across Intake, Performance, and Framework Comparison. Anchor 4 fits; not 5 because that duplication could be consolidated, not 3 because there is no padding or over-explanation.	4 / 5
Actionability	Concrete, executable guidance dominates: the autogoal bash invocation ("node .agents/skills/autogoal/scripts/create-goal-scratchpad.mjs --template major-task"), "gh issue view" / "gh pr view" fetch rules, the exact changeset auto-release blocks, and specific paths like "@docs/analysis/editor-architecture-candidates.md". Anchor 4 fits; not 5 because key contracts are delegated without content ("follow the same terse final handoff contract as task.mdc") and a few directives stay abstract ("Keep comment-back QA-focused").	4 / 5
Workflow Clarity	Intake is a 17-step numbered sequence with input classification, each execution lane is an ordered list, and the Verification section plus Success Criteria checklist provide explicit checkpoints. Anchor 4 fits; not 5 because error-recovery feedback loops (e.g. "if verification fails, fix and re-run") are never made explicit, and the flow frequently branches on conditions stated across distant sections.	4 / 5
Progressive Disclosure	There are no bundle files at all, and the entire skill is one ~345-line monolith in which lane-specific detail that belongs in reference files is inlined: the editor-framework candidate mapping, the GitHub/Linear tracked-task rules, and the changeset auto-release markup. Anchor 3 fits; not 2 because section structure is clear and well-organized rather than minimal or buried, not 4 because substantial content that should be split out is inline with no one-level-deep references.	3 / 5
	Total	15 / 20 Passed

Description

50%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 names a clear niche (heavyweight framework/library work) and its process phases, but never states when to invoke the skill, and its keywords are process jargon ("planning-first", "rigorous handoff") rather than the natural task words (architecture, migration, benchmark, proposal) a user would say. It reads as a label of the workflow's style rather than a trigger for it.

Suggestions

Append an explicit when-clause, e.g. "Use for architecture or public API redesign, framework comparison or migration, benchmarking, or RFC/proposal work" — this would lift completeness from 3 to 5 and improve trigger quality.

Replace process jargon ("planning-first research", "rigorous handoff") with the natural task vocabulary users say: "architecture", "migration", "benchmark", "RFC", "proposal".

Contrast with adjacent skills ("Not for ordinary bug fixes or one-package features") to reduce overlap risk with general research/analysis skills.

Dimension	Reasoning	Score
Specificity	Names the domain ("heavyweight framework or library tasks") and three process actions ("planning-first research, selective deep analysis, and rigorous handoff"), but these are process-framing phrases rather than concrete, verifiable capabilities like extract/fill/merge. It matches anchor 3; not 4 because the action inventory is neither specific nor comprehensive.	3 / 5
Completeness	There is a moderately clear "what" (work heavyweight tasks via research, analysis, handoff), but no "Use when..." clause or equivalent explicit trigger guidance exists anywhere in the description, which caps completeness at 3 per the judging guidelines. Not 4 because the "when" is wholly missing, not merely improvable.	3 / 5
Trigger Term Quality	Relevant keywords ("heavyweight", "framework", "library", "research", "analysis") are present, but the natural phrases a user would actually say for this skill — "architecture", "migration", "benchmark", "RFC", "proposal" — are absent. Anchor 3 fits; not 2 because more than one or two relevant domain terms appear.	3 / 5
Distinctiveness Conflict Risk	"Heavyweight framework or library tasks" is a distinguishing marker, but with no when-boundary the description overlaps broadly with any deep-research, planning, or analysis skill. Anchor 3 ("could still overlap with similar skills") is the best fit; not 4 because the overlap risk is more than minor.	3 / 5
	Total	12 / 20 Passed

Validation

81%

Warnings & errors only

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

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	13 / 16 Passed

Repository: udecode/plate
Path: .agents/skills/major-task/SKILL.md
Commit: babb3c2

Reviewed: 2 days ago

Table of Contents

Discovery Implementation Validation

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.