self-improvement

Captures learnings, errors, corrections, and feature requests to enable continuous improvement. Use when: (1) User corrects Claude ('No, that's wrong...', 'Actually...'), (2) User requests a capability that doesn't exist, (3) Claude realizes its knowledge is outdated or incorrect, (4) A better approach is discovered for a recurring task, (5) Receiving a Handoff block from self-healing (a recurring verified heal at Recurrence-Count >= 3) to distill into a memory file or new skill. For ACTIVE runtime failures where the agent needs to apply and verify a fix mid-task, use `self-healing` instead (it files HEAL- entries with proof; self-improvement promotes accumulated patterns). Also review learnings before major tasks. For CI-only/headless learning capture, use self-improvement-ci.

Quality

75%

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Passed

No known issues

Fix and improve this skill with Tessl

tessl review fix ./skills/self-improvement/SKILL.md

Quality

Content

50%

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

This skill is highly actionable with concrete templates, commands, and configurations that Claude can immediately use. However, it is severely bloated — much of the content (priority guidelines, area tags, detection trigger phrases, installation instructions, multi-agent setup) is either self-evident to Claude or could be split into reference files. The document tries to be both a quick reference and an exhaustive manual, resulting in a token-heavy monolith that undermines its own utility in a context window.

Suggestions

Cut the content by 50%+: remove installation instructions, priority/area tag tables (Claude knows these), detection trigger phrase lists, and the verbose best practices section. Keep only the Quick Reference table, logging templates, promotion rule, and hook setup.

Move the full logging templates, hook configuration details, multi-agent support, and skill extraction workflow into separate reference files (e.g., `references/templates.md`, `references/hooks-setup.md`, `references/skill-extraction.md`) and link from the main SKILL.md.

Consolidate the promotion rule — it's stated in the Simplify & Harden section, referenced in Quick Reference, and alluded to in the Promoting to Project Memory section. State it once clearly and reference it elsewhere.

Add explicit validation steps to the logging workflow (e.g., 'After appending, verify the entry ID is unique: grep LRN-YYYYMMDD .learnings/LEARNINGS.md') to strengthen workflow clarity for this file-mutation operation.

Dimension	Reasoning	Score
Conciseness	Extremely verbose at ~400+ lines. Includes installation instructions (not skill content), explains concepts Claude already knows (what corrections look like, what feature requests are), provides exhaustive tables for priority levels and area tags that are self-evident, and repeats the same promotion rule multiple times. The detection triggers section literally lists phrases like 'No, that's not right...' which Claude can recognize without being told.	1 / 3
Actionability	Provides fully concrete, copy-paste-ready templates for every entry type (learning, error, feature request), executable bash commands for setup and status checks, complete JSON configurations for hooks, and specific file paths. The logging formats and ID generation scheme are precise and immediately usable.	3 / 3
Workflow Clarity	The Quick Reference table provides good routing for different situations, and the promotion workflow has clear steps. However, the overall document lacks a clear sequential workflow — it reads more like a reference manual with many parallel sections. The Simplify & Harden ingestion workflow and skill extraction workflow have good steps but validation/verification checkpoints are mostly implicit (e.g., no explicit 'verify the entry was written correctly' step after logging).	2 / 3
Progressive Disclosure	References to external files like `references/openclaw-integration.md` and `references/hooks-setup.md` show good intent for progressive disclosure, but no bundle files were provided to verify these exist. The main SKILL.md itself is monolithic — the full logging templates, hook configurations, multi-agent setup, skill extraction workflow, and best practices are all inline when many could be split into reference files.	2 / 3
	Total	8 / 12 Passed

Description

100%

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

This is an excellent skill description that clearly defines what the skill does, when to use it with five explicit trigger scenarios, and importantly, when NOT to use it by pointing to related but distinct skills. The inclusion of natural user phrases as trigger examples and the boundary conditions with self-healing and self-improvement-ci make it highly effective for skill selection among many options.

Dimension	Reasoning	Score
Specificity	Lists multiple specific concrete actions: captures learnings, errors, corrections, and feature requests. Also specifies distilling into memory files or new skills, and reviewing learnings before major tasks.	3 / 3
Completeness	Clearly answers both 'what' (captures learnings, errors, corrections, feature requests for continuous improvement) and 'when' with an explicit numbered list of five trigger scenarios plus clear boundary conditions distinguishing it from self-healing and self-improvement-ci.	3 / 3
Trigger Term Quality	Includes natural trigger phrases users would actually say: 'No, that's wrong...', 'Actually...', capability requests, outdated knowledge, corrections. Also includes technical but relevant terms like 'Handoff block', 'self-healing', and 'HEAL- entries' for disambiguation.	3 / 3
Distinctiveness Conflict Risk	Explicitly distinguishes itself from 'self-healing' (active runtime failures) and 'self-improvement-ci' (CI-only/headless), creating clear boundaries. The specific trigger scenarios and the distinction between accumulated pattern promotion vs. active fix application make it highly distinctive.	3 / 3
	Total	12 / 12 Passed

Validation

90%

Warnings & errors only

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

Validation — 10 / 11 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (579 lines); consider splitting into references/ and linking	Warning

	Total	10 / 11 Passed

Repository: pskoett/pskoett-ai-skills
Commit: 20e64ce

Reviewed: 15 days ago

Table of Contents

Discovery Implementation Validation

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.