CtrlK
BlogDocsLog inGet started
Tessl Logo

self-improvement

Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Claude ('No, that's wrong...', 'Actually...'), (3) User requests a capability that doesn't exist, (4) An external API or tool fails, (5) Claude realizes its knowledge is outdated or incorrect, (6) A better approach is discovered for a recurring task. Also review learnings before major tasks.

68

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Critical

Do not install without reviewing

SKILL.md
Quality
Evals
Security

Quality

Content

77%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A highly actionable, well-structured skill body with concrete templates, commands, and sequenced workflows including verification steps. Its main weaknesses are verbosity from repeated promotion-target tables and a monolithic SKILL.md that underuses the available reference files.

Suggestions

De-duplicate the Promotion Targets material: keep one canonical table and reference it from the other sections instead of repeating it three times.

Move the full Logging Format templates and Multi-Agent setup details into references/examples.md and references/openclaw-integration.md, leaving SKILL.md as a concise overview that links out, to lift progressive_disclosure to level 3.

Trim the Detection Triggers section since the same triggers already appear in the frontmatter description.

DimensionReasoningScore

Conciseness

The body is mostly useful but heavily redundant: the Promotion Targets table recurs in the Quick Reference, the OpenClaw section, and the 'Promoting to Project Memory' section, and the Detection Triggers repeat the frontmatter triggers. It could be tightened substantially, but it is not level-1 fluff explaining concepts Claude already knows. It is not level 3 because the repetition and length mean not every token earns its place.

2 / 3

Actionability

It provides copy-paste-ready entry templates, concrete shell commands (mkdir, grep one-liners, hook JSON config, extract-skill.sh invocations), and explicit ID/status formats. This matches the level-3 'fully executable, copy-paste ready' anchor and is well above the level-2 pseudocode anchor.

3 / 3

Workflow Clarity

Multi-step processes (Self-Improvement Workflow, Extraction Workflow, Simplify & Harden Ingestion) are clearly numbered, and the extraction workflow includes an explicit 'Verify: Read skill in fresh session' checkpoint plus a quality-gates checklist. It is not level 2 because validation/verification steps are present where they matter rather than only implicit.

3 / 3

Progressive Disclosure

Real, one-level-deep references are clearly signaled ('See references/openclaw-integration.md', 'See references/hooks-setup.md', assets/SKILL-TEMPLATE.md, scripts/*), but SKILL.md itself is a ~640-line monolith with full logging templates and agent-specific setup inline that could live in the existing reference files. It is not level 3 because content that should be split out remains inline; it is above level 1 because references are real and clearly signaled rather than deeply nested.

2 / 3

Total

10

/

12

Passed

Description

90%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description with explicit, natural-language triggers and a clear what/when structure written in correct third-person voice. Its only weakness is that the named actions are somewhat abstract rather than enumerating concrete operations.

Suggestions

Replace abstract verbs with concrete actions, e.g. 'Logs learnings, errors, and corrections to markdown files and promotes high-value entries to CLAUDE.md/AGENTS.md' to lift specificity to level 3.

DimensionReasoningScore

Specificity

The description names the domain and actions ('Captures learnings, errors, and corrections to enable continuous improvement') but the verbs are abstract rather than a list of multiple concrete actions like 'log to markdown files, promote to project memory'. It is not the level-3 'multiple specific concrete actions' anchor and clearly above the level-1 vague anchor.

2 / 3

Completeness

It states what the skill does ('Captures learnings, errors, and corrections...') and gives an explicit 'Use when: (1)...(6)' trigger list plus 'Also review learnings before major tasks', answering both what and when. It is not level 2 because the 'when' is explicit, not merely implied.

3 / 3

Trigger Term Quality

It quotes natural phrases users actually say ('No, that's wrong...', 'Actually...') plus concrete situations ('command or operation fails unexpectedly', 'external API or tool fails'), giving good coverage of natural trigger terms. It is not level 2 because it goes well beyond a single keyword to a broad set of realistic utterances.

3 / 3

Distinctiveness Conflict Risk

The niche (capturing learnings/errors/corrections for continuous improvement) is distinct and the triggers (user corrections, command failures, missing features, outdated knowledge) are unlikely to fire for unrelated skills. It is above level 2 because the trigger set is specific rather than a generic 'works with files' overlap.

3 / 3

Total

11

/

12

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

skill_md_line_count

SKILL.md is long (648 lines); consider splitting into references/ and linking

Warning

Total

15

/

16

Passed

Repository
wanikua/danghuangshang
Reviewed

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.