Analyze team workflows to identify missing skills, agent behavior gaps, or repeated corrections that indicate need for new skills. Use when analyzing what skills would help, identifying why agents keep making the same mistakes, conducting skill gap analysis, or when user asks what skills would help, why do agents keep doing X wrong, skill audit. Triggers on repeated corrections, agent failures, workflow inconsistencies, skill gap, missing skill, agents keep.
How this skill is triggered — by the user, by Claude, or both
Slash command
/gutt-claude-code-plugin:skills-discoveryThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
**Announce:** "Analyzing workflows for skill gaps..."
Announce: "Analyzing workflows for skill gaps..."
Identify opportunities for new skills based on repeated patterns.
| Trigger | Action |
|---|---|
| Same correction made 3+ times | Skill to prevent error |
| Complex workflow repeated | Skill to standardize |
| Agent confusion patterns | Skill to clarify |
| New tool/integration | Skill to document usage |
# Find repeated corrections
fetch_lessons_learned(query="correction mistake error", max_results=20)
# Find behavioral signals
search_memory_nodes(query="correction pattern", entity="BehavioralSignal", max_nodes=10)
# Find decisions that need enforcement
search_memory_facts(query="decision standard process", max_facts=15)
Group findings by:
| Category | Indicates |
|---|---|
| Repeated "Negation" lessons | Skill to prevent misunderstanding |
| Repeated "Replacement" lessons | Skill to enforce correct approach |
| Multiple decisions in one area | Skill to consolidate guidance |
| Tool-specific errors | Skill to document tool usage |
## Proposed Skill: [name]
### Evidence
- Lesson (uuid: xxx): [pattern]
- Lesson (uuid: xxx): [pattern]
### Problem
[What keeps going wrong]
### Solution
Skill that:
- [What it would do]
- [When it would trigger]
- [What it would prevent]
### Estimated Impact
[How much time/errors this would save]
add_memory(
name="Skill Gap: [area]",
episode_body="Identified need for [skill type] based on [evidence]. Proposed: [brief description].",
source="text"
)
name: prevent-[error-type]
description: Prevent [common error] by [mechanism]. Triggers on [patterns].
name: [workflow]-standard
description: Standardize [workflow] using established patterns. Triggers on [task types].
name: [tool]-usage
description: Correct usage of [tool] including common pitfalls. Triggers on [tool name], [common errors].
Report: "skills-discovery: VERIFIED - Found [X patterns], proposed [Y skills]"
npx claudepluginhub ibrain-bvba/gutt-claude-code-plugin --plugin gutt-claude-code-pluginAnalyzes skill outcome logs and user corrections, then proposes self-improvements to skill behavior and description hygiene.
Mines coding-agent session history, transcripts, and memories to discover recurring workflows and draft new Agent Skills from real usage evidence.
Detects repeated workflow patterns from AI session transcripts and suggests reusable skills to automate them. Use when optimizing workflows.