CtrlK
BlogDocsLog inGet started
Tessl Logo

self-improving-agent

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User requests a capability that doesn't exist, (4) An external API or tool fails, (5) Claude realizes its knowledge is outdated or incorrect, (6) A better approach is discovered for a recurring task.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.

76

Quality

95%

Does it follow best practices?

Impact

No eval scenarios have been run

SecuritybySnyk

Passed

No known issues

SKILL.md
Quality
Evals
Security

Quality

Content

100%

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

A well-structured, instruction-only skill that provides concrete templates, clear file locations, and a sensible workflow with safety boundaries. It is lean and actionable, with only minor redundancy across the record-type templates.

Suggestions

Define the shared record header (Logged, Priority, Status, Area) once and have each record type reference it, removing the repeated boilerplate across the three templates.

Fold the 'verify before marking resolved' rule into the Workflow as an explicit step so the validation gate appears in the process sequence rather than only in Safety Boundaries.

DimensionReasoningScore

Conciseness

The body is lean and template-driven with no explanation of concepts Claude already knows; it assumes competence and avoids filler, though the three record templates repeat shared metadata headers that could be factored.

3 / 3

Actionability

It gives concrete file paths (.learnings/LEARNINGS.md), exact ID formats (LRN-YYYYMMDD-XXX), enumerated Priority/Area values, and copy-paste-ready record templates, so the guidance is specific and complete rather than abstract.

3 / 3

Workflow Clarity

A numbered four-step Workflow sequences the process, and the 'Do not mark an entry resolved unless the fix was actually verified' rule supplies an explicit verification gate; for this low-risk logging skill the sequence is clear and sufficient.

3 / 3

Progressive Disclosure

No bundle files exist, so the single self-contained SKILL.md is organized into clear sections (When to Use, Storage, Record Types, Workflow, Promotion Rules, Safety Boundaries) with no nested references, fitting the well-organized single-file carve-out.

3 / 3

Total

12

/

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 that pairs a clear purpose statement with an explicit, scenario-based 'Use when' clause written in third person. The only weakness is that the 'what it does' actions are somewhat abstract categories rather than concrete operations.

Suggestions

Sharpen the capability statement with concrete verbs, e.g. 'Appends structured learning, error, and feature-request notes to a local .learnings/ knowledge base' to lift specificity.

Consider adding the trigger keyword 'learnings' or 'past lessons' near the 'Use when' list so a user asking to 'review past learnings' lands on this skill.

DimensionReasoningScore

Specificity

The description names the captured categories ('learnings, errors, and corrections') and the goal ('enable continuous improvement'), but the actions are general ('Captures', 'review') rather than multiple concrete granular actions, matching the 'names domain and some actions' anchor.

2 / 3

Completeness

It explicitly answers both what ('Captures learnings, errors, and corrections to enable continuous improvement') and when via an explicit, numbered 'Use when' clause plus 'review learnings before major tasks', so both halves are clearly present.

3 / 3

Trigger Term Quality

The numbered 'Use when' clauses use natural scenario phrasings a user would actually encounter ('command or operation fails unexpectedly', 'User corrects Claude', 'a capability that doesn't exist', 'outdated or incorrect'), giving good coverage of natural trigger terms.

3 / 3

Distinctiveness Conflict Risk

The self-improvement/learnings-log niche with distinct situational triggers (corrections, outdated knowledge, missing capabilities) is unlikely to fire for unrelated skills, fitting the clear-niche anchor.

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

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository
jdrhyne/agent-skills
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.