CtrlK
BlogDocsLog inGet started
Tessl Logo

getlarge/legreffier-explore

Systematic diary exploration: discover tags, entry distribution, coverage gaps, agent mistakes, and compile recipes

86

1.06x
Quality

90%

Does it follow best practices?

Impact

81%

1.06x

Average score across 5 eval scenarios

SecuritybySnyk

Advisory

Suggest reviewing before use

Overview
Quality
Evals
Security
Files

Evaluation results

57%

5%

Diary Health Report Generator

Full exploration report and recovery protocol

Criteria
Without context
With context

All report sections

83%

83%

Entry type reflection

0%

0%

Exploration tags

50%

50%

Importance 6

0%

0%

Recovery detection

66%

100%

Phase resumption

100%

90%

Incident-antipattern relations

37%

87%

Decision-commit relations

37%

25%

Repeated incident relations

12%

62%

Header fields

62%

50%

Severity grouping in report

33%

16%

Ordered phases

100%

66%

92%

Diary Inventory Analyzer

Diary inventory and tag namespace discovery

Criteria
Without context
With context

Pagination function

100%

100%

Batch size 50

0%

0%

Entry type counts

100%

100%

Tag frequency

100%

100%

Tag namespace grouping

100%

100%

Namespace discovery not hardcoded

100%

100%

Importance histogram

100%

100%

Temporal range

100%

100%

Sample report generated

100%

100%

Tag tree format

100%

100%

Total entry count

100%

100%

97%

5%

Incident Triage Tool for Diary Entries

Incident triage and episodic analysis

Criteria
Without context
With context

Tag-first search

100%

100%

Fallback search

100%

100%

Structured extraction

100%

100%

Severity classification

100%

100%

Subsystem inference

100%

100%

Subsystem grouping

100%

100%

Training candidates

100%

100%

Non-episodic filtering

50%

100%

Untagged incident detection

100%

100%

Preventive context field

100%

100%

Severity in report

50%

62%

72%

5%

Commit Quality Analyzer

Commit pattern analysis and anti-pattern detection

Criteria
Without context
With context

Procedural tag search

0%

0%

Fallback search

0%

0%

Tag frequency within procedural

100%

100%

Branch grouping

100%

100%

Double-prefix detection

100%

100%

Missing scope detection

0%

100%

Missing branch detection

100%

100%

Catch-all tag detection

75%

75%

Broad entry detection

62%

0%

Non-procedural filtering

100%

100%

Report has anti-patterns section

100%

100%

91%

12%

Context Pack Recipe Generator

Coverage gap detection and compile recipe design

Criteria
Without context
With context

Recipe YAML fields

58%

58%

Recipes use existing tags

100%

100%

Coverage gap detection

100%

100%

Learn trace analysis

60%

80%

Noise source identification

80%

100%

Noise in exclude_tags

50%

75%

Multiple recipes

100%

100%

Rationale per recipe

37%

100%

Weight parameters

87%

100%

Token budget specified

100%

100%

Gap evidence

100%

100%

Evaluated
Agent
Claude Code
Model
Claude Sonnet 4.6

Table of Contents