rt-ica
Reverse Thinking - Information Completeness Assessment. Mandatory pre-planning checkpoint that blocks planning until prerequisites are verified. Use when receiving specs, PRDs, tickets, RFCs, architecture designs, or any multi-step engineering task. Integrates with CoVe-style planning pipelines. Invoke BEFORE creating plans, delegating to agents, or defining acceptance criteria.
85
Quality
79%
Does it follow best practices?
Impact
94%
1.36xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./.claude/skills/rt-ica/SKILL.mdEvaluation results
99%
29%Notification Service: Initial Planning Request
BLOCKED decision with structured missing-input questions
No plan produced
100%
100%
Goal statement present
100%
100%
Success output defined
100%
100%
Conditions enumerated
50%
100%
AVAILABLE/DERIVABLE/MISSING statuses
0%
100%
MISSING conditions identified
40%
100%
BLOCKED decision stated
66%
100%
Questions only for missing inputs
90%
100%
Questions categorized
100%
100%
Questions ordered by criticality
75%
100%
Constrained question format
0%
75%
No silent assumptions
100%
100%
Without context: $0.1677 · 1m 5s · 8 turns · 12 in / 2,719 out tokens
With context: $0.4333 · 1m 48s · 21 turns · 278 in / 5,251 out tokens
87%
21%Inventory Sync Service: Engineering Plan
APPROVED decision with complete plan structure
RT-ICA before plan
100%
100%
Goal reconstruction present
50%
100%
Conditions with required info
83%
100%
Availability statuses used
0%
100%
AVAILABLE conditions cited
16%
100%
APPROVED decision stated
0%
0%
DERIVABLE items as assumptions
37%
62%
Plan: Workstreams section
100%
100%
Plan: Guardrails section
83%
100%
Plan: Acceptance Criteria section
100%
100%
Plan: Risk Register section
100%
100%
Plan: Dependencies section
100%
100%
Plan: Verification Plan section
100%
100%
Plan: Change Management section
83%
100%
Conditions enforced as guardrails
100%
100%
Without context: $0.2513 · 2m 11s · 8 turns · 13 in / 6,229 out tokens
With context: $0.7054 · 3m 52s · 21 turns · 184 in / 11,955 out tokens
97%
24%Search Index Migration: Proceed with Assumptions
Assumption-based planning with risk notes and early validation tasks
RT-ICA analysis present
83%
100%
Conditions with statuses
25%
100%
MISSING conditions identified
75%
100%
BLOCKED decision stated
20%
90%
Explicit assumption per missing condition
92%
100%
Risk note per assumption
92%
100%
Early validation tasks
92%
100%
No silent assumptions
80%
80%
Plan proceeds to implementation
100%
100%
DERIVABLE items handled
37%
100%
Without context: $0.2252 · 1m 38s · 8 turns · 11 in / 4,546 out tokens
With context: $0.4617 · 2m 53s · 16 turns · 19 in / 7,621 out tokens
- Repository
- Jamie-BitFlight/claude_skills
- Commit
5667e97
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6
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.