CtrlK
BlogDocsLog inGet started
Tessl Logo

openspec-apply-change

Implement tasks from an OpenSpec change. Use when the user wants to start implementing, continue implementation, or work through tasks.

76

1.06x
Quality

67%

Does it follow best practices?

Impact

87%

1.06x

Average score across 3 eval scenarios

SecuritybySnyk

Advisory

Suggest reviewing before use

Optimize this skill with Tessl

npx tessl skill review --optimize ./.github/skills/openspec-apply-change/SKILL.md
SKILL.md
Quality
Evals
Security

Evaluation results

99%

11%

Implementing a Feature Change with OpenSpec

Full apply workflow CLI sequence and progress tracking

Criteria
Without context
With context

openspec status command

100%

100%

openspec instructions apply command

100%

100%

contextFiles used from CLI output

90%

90%

Progress display before implementation

100%

100%

Checkbox updated per task

100%

100%

Task-by-task progression

100%

100%

Change announcement

87%

100%

Completion summary format

87%

100%

Archive suggestion

50%

100%

Minimal scoped changes

100%

100%

Session tasks listed

50%

100%

No assumed file names

87%

100%

85%

-7%

Continuing Development After a Partial Implementation

Resume partial implementation, progress tracking and checkbox updates

Criteria
Without context
With context

Schema identified

100%

100%

Starting progress shown

100%

100%

contextFiles read from CLI output

30%

20%

Only remaining tasks implemented

90%

40%

Region filter added

100%

100%

Summary export endpoint added

100%

100%

README_export.md created

100%

100%

Checkboxes updated

100%

100%

Task-by-task documentation

100%

100%

Final progress in report

100%

100%

Archive suggestion in report

100%

100%

Minimal changes

100%

87%

78%

10%

Feature Implementation with a Missing Dependency

Blocked state handling, pause output format, do not guess

Criteria
Without context
With context

Schema identified

100%

100%

CLI outputs consulted

62%

62%

contextFiles from CLI output

50%

50%

Task 1 completed

100%

37%

Task 1 checkbox updated

50%

100%

Did not guess config

100%

100%

Did not implement task 2

87%

100%

Paused state output format

25%

50%

Progress count in report

25%

100%

Blocker description

100%

100%

Numbered options provided

100%

100%

User prompt at end

27%

45%

Repository
arm/mlia
Evaluated
Agent
Claude Code
Model
Claude Sonnet 4.6

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.