Name: autogoal
Rating: 52.8 (1 reviews)
Author: udecode

autogoal

Create, verify, repair, and close durable Codex goals with measurable outcomes, evidence gates, plan templates, blocker handling, completion audits, and goal-backed workflow repair.

Quality

60%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./templates/plate-template/.agents/skills/autogoal/SKILL.md

The canonical home for this skill is autogoal in udecode/plate

Quality

Content

62%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 exceptionally actionable and the workflows are clearly sequenced with layered validation gates, but it pays for that with heavy verbosity: rules are restated across sections and large reference-grade material (plan schema, template quality bar, repair matrix) is inlined in a ~1,280-line monolith that never points to the real bundle files that already exist. Moving the reference-grade material into references/ and deduplicating the repeated plan-fill rules would raise both conciseness and progressive disclosure without losing actionability.

Suggestions

Move the required goal-plan section schema, the template input checklist / quality bar, and the repair scope matrix into reference files under references/ and link to them from the body, instead of inlining ~400 lines of reference-grade material.

Deduplicate the plan-editing rules — "fill the generated plan, do not hand-replace it, mark N/A with reason" is stated nearly identically in Template Composition, Goal Plan, and Linked Plan Trees; state it once and cross-reference.

Trim or merge overlapping sections (Budget Handling vs Output Budget Discipline, Completion Gate Policy vs Completion Rules, Template Init vs Template Composition) to cut the body to a lean core, since nearly every rule is currently stated two to three times.

Dimension	Reasoning	Score
Conciseness	The ~1,280-line body repeats the same rules in multiple sections — plan-fill/"do not hand-replace the generated plan" rules appear in Template Composition, Goal Plan, and Linked Plan Trees; template paths are listed twice; "Autoreview is never a universal gate" appears twice — matching the 'noticeably verbose; several unnecessary explanations or padded sections' anchor. It is not 1 because the content is domain-specific policy rather than explanation of concepts Claude already knows; it is not 3 because the duplication is pervasive, not occasional.	2 / 5
Actionability	Concrete, copy-paste-ready commands appear throughout ("node .agents/skills/autogoal/scripts/create-goal-scratchpad.mjs --template task --with docs", "check-complete.mjs <docs/plans/path>", "init-templates.mjs") plus exact objective shapes, plan section formats, and template/pack names that all match real bundle files, matching the 'mostly executable guidance with minor gaps' anchor. It is not 5 because a substantial share of the body is judgment-call policy prose (repair scope matrices, conflict protocol classifications) that instructs by rule rather than by executable example.	4 / 5
Workflow Clarity	Multi-step processes are explicitly sequenced with validation checkpoints and feedback loops: the 15-step Start Workflow, the Completion Gate Policy requiring check-complete.mjs as the final mechanical gate before update_goal, the repair workflow's "prove the repair" verification steps, and blocked-vs-complete rules with explicit do/do-not lists — matching the 'clear sequence with explicit validation steps; feedback loops for error recovery; checklists' anchor. It is not 4 because validation is not merely present but layered (per-step gates plus a final mechanical checker plus recursive child-plan validation).	5 / 5
Progressive Disclosure	The bundle structure is real and consistent — every referenced script (create-goal-scratchpad.mjs, check-complete.mjs, init-templates.mjs, create-goal-template.mjs) and every named template/pack (task, docs, major-task, goal-repair, packs/agent-native, browser, package-api, performance-observability) exists on disk — but the SKILL.md body inlines material that belongs in reference files (the full required goal-plan section schema, the template quality bar and template input checklist, the repair scope matrix), matching the 'some structure but content that should be separate is inline' anchor. It is above 2 because section headers are clear and the scripts are genuinely used, but below 4 because nothing in the body points the reader to the existing assets/templates or other reference files.	3 / 5
	Total	14 / 20 Passed

Description

58%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 gives a concrete, multi-action picture of what the skill does, but it reads as an internal capability inventory rather than a trigger-oriented description: it lacks any 'use when' guidance and relies on jargon instead of phrases a user would naturally say. Adding explicit trigger conditions and natural-language keywords would lift the two heaviest-weighted dimensions.

Suggestions

Append a trigger clause such as: "Use when the user asks to set a goal, keep working until a verifiable end state, run long-running autonomous work, or says 'keep going until it works'."

Replace jargon-heavy nouns ("evidence gates", "completion audits") with one or two natural user-facing phrases so trigger matching works on what users actually type.

State the 'when' explicitly even if brief — e.g., "Use for debugging loops, migrations, flaky-test hunts, or any work with an auditable finish line" — to move completeness from 3 to 4-5.

Dimension	Reasoning	Score
Specificity	The description lists several concrete actions — "Create, verify, repair, and close" — tied to specific artifacts ("measurable outcomes, evidence gates, plan templates, blocker handling, completion audits"), matching the 'several specific actions; minor gaps' anchor. It falls short of 5 because the action nouns are domain jargon stacks rather than fully unpacked capabilities, and it exceeds 3 since multiple distinct actions are named, not just 1-2.	4 / 5
Completeness	The 'what' is clearly stated (create, verify, repair, close goals with measurable outcomes and evidence gates) but there is no 'Use when...' clause or equivalent trigger guidance, so per the judging guideline completeness is capped at 3. It is above 2 because the 'what' is concrete and multi-part, and it cannot be 4-5 because 'when' is entirely missing rather than merely implicit.	3 / 5
Trigger Term Quality	Relevant keywords exist ("goals", "repair", "plan templates", "completion audits") but the phrasing is internal jargon ("durable Codex goals", "evidence gates", "goal-backed workflow repair") rather than natural user phrases like "set a goal", "keep working until done", or "long-running task", matching the 'some relevant keywords but missing common variations' anchor. It is above 2 because the domain terms are on-topic, but below 4 because everyday synonyms a user would actually say are absent.	3 / 5
Distinctiveness Conflict Risk	"durable Codex goals" with "evidence gates" and "completion audits" carves a fairly distinct lifecycle niche that would rarely collide with unrelated skills, matching the 'mostly distinct; minor overlap risk' anchor. It is not 5 because goal/planning/repair vocabulary overlaps with generic planning and workflow-repair skills, and not 3 because the specific goal-tool framing is recognizably unique.	4 / 5
	Total	14 / 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
skill_md_line_count	SKILL.md is long (1283 lines); consider splitting into references/ and linking	Warning
relative_links	Relative link issues: 2 missing, 2 deeper-than-1-level	Warning
referenced_paths_exist	Referenced path issues: 2 missing	Warning

	Total	13 / 16 Passed

Repository: udecode/plate
Path: templates/plate-template/.agents/skills/autogoal/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.