continuous-learning-v2

Instinct-based learning system that observes sessions via hooks, creates atomic instincts with confidence scoring, and evolves them into skills/commands/agents.

Quality

52%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./docs/zh-TW/skills/continuous-learning-v2/SKILL.md

Quality

Content

46%Weight 40%Scale 1-5

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

The body is well-organized with concrete setup steps, but it is held back by padded conceptual/promotional sections, inlined configuration that belongs in separate files, and — most seriously — references to hooks, scripts, and commands that are not actually shipped in the bundle, leaving the core learning workflow unexecutable. The setup half is actionable; the operate-and-evolve half is aspirational documentation.

Suggestions

Ship the referenced bundle files (hooks/hooks.json, hooks/observe.sh, agents/start-observer.sh, the four command definitions) or remove/inline their behavior so every instruction is executable as delivered.

Trim the v1/v2 comparison table, the "Why Hooks vs Skills" rationale, and the Skill Creator integration section — they explain motivation Claude does not need — and move the full config.json block to a separate reference file.

Add validation checkpoints to the workflow: after enabling hooks, verify observations.jsonl is receiving entries; after running the observer, confirm instincts were created before running /evolve.

Dimension	Reasoning	Score
Conciseness	Concrete sections (hook JSON, mkdir setup, config.json, instinct YAML format) are efficient, but the v1/v2 comparison table, the "為何 Hooks vs Skills 用於觀察？" conceptual rationale, Skill Creator integration promotion, and external-link sections add padding beyond minor over-explanation. Anchor 3 fits better than 4, which requires only minor trimmable instances.	3 / 5
Actionability	Hook JSON, mkdir commands, and the instinct YAML example are executable, but the core actions are undefined: /evolve, /instinct-status, /instinct-export, /instinct-import have only one-line descriptions with no implementation guidance, and the referenced scripts (hooks/observe.sh, agents/start-observer.sh) are not in the bundle. "Some concrete guidance but incomplete; missing key details" matches anchor 3.	3 / 5
Workflow Clarity	The quick start provides a clear 1-2-3 sequence, but there are no validation checkpoints: nothing verifies hooks actually fire, that observations.jsonl is being written, or how to recover when the observer misbehaves. Anchor 3 (sequence present, checkpoints missing) fits better than 4.	3 / 5
Progressive Disclosure	The ~260-line body is monolithic: the full hooks JSON, config.json, and command set are inlined rather than split out, and every referenced bundle artifact (hooks/hooks.json, hooks/observe.sh, agents/start-observer.sh, the four commands) does not exist in the shipped bundle — there are no references/, scripts/, or assets/ directories at all. This matches anchor 2: content that clearly belongs in separate files is inlined and references point to nothing.	2 / 5
	Total	11 / 20 Passed

Description

58%Weight 40%Scale 1-5

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

The description states a clear, multi-action "what" with a distinctive niche, but it entirely lacks a "when" clause and uses coined jargon ("instincts") instead of natural trigger terms a user would actually say. Adding an explicit "Use when..." sentence with user-natural phrasing would lift both completeness and trigger-term quality.

Suggestions

Append an explicit trigger clause, e.g. "Use when the user asks Claude to learn from their corrections, remember their preferences, or build reusable skills from session history."

Replace or supplement coined jargon ("instinct-based", "atomic instincts") with natural synonyms users would say, such as "learned preferences", "patterns", "personalization".

Mention the export/import capability in the description so the "what" coverage is comprehensive.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions — "observes sessions via hooks", "creates atomic instincts with confidence scoring", "evolves them into skills/commands/agents" — with only minor coverage gaps (export/import is unmentioned). Not a 5 because coverage is not comprehensive; not a 3 because it names the domain plus more than 1-2 specific actions.	4 / 5
Completeness	The "what" is clear (observe sessions, create instincts, evolve them), but there is no "Use when..." clause or equivalent explicit trigger guidance, which caps completeness at 3 per the judging guidelines. Would need explicit trigger phrases to reach 4-5.	3 / 5
Trigger Term Quality	Contains relevant keywords ("learning", "hooks", "skills/commands/agents") but leans on invented jargon ("instinct-based", "atomic instincts") and misses natural user phrases like "learn from my corrections" or "remember my preferences". Anchor 3 (some relevant keywords, missing common variations) fits better than 4.	3 / 5
Distinctiveness Conflict Risk	Defines a clear niche — the instinct lifecycle (observe → instinct → evolve with confidence scoring) — that is mostly distinct with only minor overlap risk against general memory/learning skills. Not a 5 because "skills/commands/agents" and "learning" are broad terms that could overlap with related skills.	4 / 5
	Total	14 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: affaan-m/ECC
Path: docs/zh-TW/skills/continuous-learning-v2/SKILL.md
Commit: c70874f

Reviewed: 1 day 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.