# Agent task execution on SkillsBench 11 domains Source: [Agent task execution on SkillsBench 11 domains]() Collection index: [llms.txt]() Data updated: 2026-06-08T06:42:07.299Z Evaluation of general agent task execution performance across the 11 specified domains within the SkillsBench framework, reporting overall success and domain-specific scores. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Agent task execution | | Dataset (including any stated split) | SkillsBench 11 domains | | Metric | Overall Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 16 of 16 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Overall Score](): Higher is better; unit: PERCENT. - [Software Domain Score](): Higher is better; unit: PERCENT. - [Office Domain Score](): Higher is better; unit: PERCENT. - [Science Domain Score](): Higher is better; unit: PERCENT. - [Media Domain Score](): Higher is better; unit: PERCENT. - [Cybersecurity Domain Score](): Higher is better; unit: PERCENT. - [Finance Domain Score](): Higher is better; unit: PERCENT. - [Robotics Domain Score](): Higher is better; unit: PERCENT. - [Energy Domain Score](): Higher is better; unit: PERCENT. - [Manufacturing Domain Score](): Higher is better; unit: PERCENT. - [Health Domain Score](): Higher is better; unit: PERCENT. - [Math Domain Score](): Higher is better; unit: PERCENT. - [Improvement vs No Skill](): Higher is better; unit: SCALAR. ## Recorded results | Method | Overall Score | Result date | Report paper | | --- | --- | --- | --- | | Human | 44.8 | Not recorded | 6a264c8e90966ade919be9f1 | | Human | 44.5 | Not recorded | 6a264c8e90966ade919be9f1 | | OPENSKILL | 43.6 | Not recorded | 6a264c8e90966ade919be9f1 | | OPENSKILL | 42.1 | Not recorded | 6a264c8e90966ade919be9f1 | | Skill-Creator | 34.7 | Not recorded | 6a264c8e90966ade919be9f1 | | CoT | 33.3 | Not recorded | 6a264c8e90966ade919be9f1 | | Self-Gen | 32.2 | Not recorded | 6a264c8e90966ade919be9f1 | | Memento | 30.1 | Not recorded | 6a264c8e90966ade919be9f1 | | Skill-Creator | 29.2 | Not recorded | 6a264c8e90966ade919be9f1 | | No Skill | 25.5 | Not recorded | 6a264c8e90966ade919be9f1 | | No Skill | 25 | Not recorded | 6a264c8e90966ade919be9f1 | | AutoSkill | 24.7 | Not recorded | 6a264c8e90966ade919be9f1 | | Self-Gen | 23.9 | Not recorded | 6a264c8e90966ade919be9f1 | | CoT | 23.9 | Not recorded | 6a264c8e90966ade919be9f1 | | Memento | 15.6 | Not recorded | 6a264c8e90966ade919be9f1 | | AutoSkill | 11.2 | Not recorded | 6a264c8e90966ade919be9f1 | ### Human Overall Score: 44.8. Result ID: 6a2663f6fbc373592c227dda. [Reporting paper]() Recorded evaluation configuration: {"Target Agent":"GPT 5.2 (Codex)"}. This record does not assert independent reproduction. ### Human Overall Score: 44.5. Result ID: 6a2663f6fbc373592c227dd2. [Reporting paper]() Recorded evaluation configuration: {"Target Agent":"Opus 4.6 (Claude Code)"}. This record does not assert independent reproduction. ### OPENSKILL Overall Score: 43.6. Result ID: 6a2663f6fbc373592c227dd1. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Target Agent":"Opus 4.6 (Claude Code)"}. This record does not assert independent reproduction. ### OPENSKILL Overall Score: 42.1. Result ID: 6a2663f6fbc373592c227dd9. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Target Agent":"GPT 5.2 (Codex)"}. This record does not assert independent reproduction. ### Skill-Creator Overall Score: 34.7. Result ID: 6a2663f6fbc373592c227dce. [Reporting paper]() Recorded evaluation configuration: {"Target Agent":"Opus 4.6 (Claude Code)"}. This record does not assert independent reproduction. ### CoT Overall Score: 33.3. Result ID: 6a2663f6fbc373592c227dd5. [Reporting paper]() Recorded evaluation configuration: {"Target Agent":"GPT 5.2 (Codex)"}. This record does not assert independent reproduction. ### Self-Gen Overall Score: 32.2. Result ID: 6a2663f6fbc373592c227dd4. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Target Agent":"GPT 5.2 (Codex)"}. This record does not assert independent reproduction. ### Memento Overall Score: 30.1. Result ID: 6a2663f6fbc373592c227dd0. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Target Agent":"Opus 4.6 (Claude Code)"}. This record does not assert independent reproduction. ### Skill-Creator Overall Score: 29.2. Result ID: 6a2663f6fbc373592c227dd6. [Reporting paper]() Recorded evaluation configuration: {"Target Agent":"GPT 5.2 (Codex)"}. This record does not assert independent reproduction. ### No Skill Overall Score: 25.5. Result ID: 6a2663f6fbc373592c227dcb. [Reporting paper]() Recorded evaluation configuration: {"Target Agent":"Opus 4.6 (Claude Code)"}. This record does not assert independent reproduction. ### No Skill Overall Score: 25. Result ID: 6a2663f6fbc373592c227dd3. [Reporting paper]() Recorded evaluation configuration: {"Target Agent":"GPT 5.2 (Codex)"}. This record does not assert independent reproduction. ### AutoSkill Overall Score: 24.7. Result ID: 6a2663f6fbc373592c227dcf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Agent":"Opus 4.6 (Claude Code)"}. This record does not assert independent reproduction. ### Self-Gen Overall Score: 23.9. Result ID: 6a2663f6fbc373592c227dcc. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Target Agent":"Opus 4.6 (Claude Code)"}. This record does not assert independent reproduction. ### CoT Overall Score: 23.9. Result ID: 6a2663f6fbc373592c227dcd. [Reporting paper]() Recorded evaluation configuration: {"Target Agent":"Opus 4.6 (Claude Code)"}. This record does not assert independent reproduction. ### Memento Overall Score: 15.6. Result ID: 6a2663f6fbc373592c227dd8. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Target Agent":"GPT 5.2 (Codex)"}. This record does not assert independent reproduction. ### AutoSkill Overall Score: 11.2. Result ID: 6a2663f6fbc373592c227dd7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Agent":"GPT 5.2 (Codex)"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.