Module: Clacky::Agent::SkillReflector

Included in:
Clacky::Agent
Defined in:
lib/clacky/agent/skill_reflector.rb

Overview

Scenario 2: Reflect on skill execution and suggest improvements.

After a skill completes, forks a subagent to analyze:

- Were instructions clear enough?
- Any missing edge cases?
- Any improvements needed?

If the LLM identifies concrete improvements, it invokes skill-creator to update the skill.

Constant Summary collapse

MIN_SKILL_ITERATIONS =

Minimum iterations for a skill execution to warrant reflection. This counts iterations within the skill execution only, not session-cumulative.

5

Instance Method Summary collapse

Instance Method Details

#maybe_reflect_on_skill ⇒ Object

Check if we should reflect on the skill that just executed Called from SkillEvolution#run_skill_evolution_hooks



21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
 
# File 'lib/clacky/agent/skill_reflector.rb', line 21

def maybe_reflect_on_skill
  return unless @skill_execution_context

  # Only reflect on skills that the user explicitly invoked via slash command.
  # Skills triggered by the LLM itself (e.g. as part of a broader task) or
  # platform-management skills invoked incidentally should not be reflected on.
  return unless @skill_execution_context[:slash_command]

  # Skip default and brand skills — they are system-owned and should not be
  # auto-improved by the evolution system.
  source = @skill_execution_context[:source]
  return if source == :default || source == :brand

  skill_name = @skill_execution_context[:skill_name]
  start_iteration = @skill_execution_context[:start_iteration]
  
  # Calculate iterations within the skill execution (not session-cumulative)
  iterations = @iterations - start_iteration

  # Only reflect if the skill actually ran for a meaningful number of iterations
  return if iterations < MIN_SKILL_ITERATIONS

  # Fork an isolated subagent to reflect + improve — does NOT touch main history
  @ui&.show_info("Reflecting on skill execution: #{skill_name}")
  subagent = fork_subagent
  subagent.run(build_skill_reflection_prompt(skill_name))

  # Clear the context so we don't reflect again
  @skill_execution_context = nil
end