Skip to content
Back to skills

Controlflow Assumption Verifier

ASecurity

Use when a saved plan may contain hidden assumptions disguised as facts, especially for medium or large tasks, unresolved high-risk plans, or plans that mention files, APIs, versions, integrations, tests, or concurrency patterns that should be verified against the repository before execution.

  • 13 stars
  • 0 votes
  • 0 copies
  • 6 views
  • Added May 27, 2026
data-aiapisecurity

Works with

  • api

Security analysis

A100/100

Pro scans all 2 files and shows the line behind each finding

Scanned September 6, 2026

npx -y skills add Smithbox-ai/ControlFlow --skill controlflow-assumption-verifier --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Controlflow Assumption Verifier?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Controlflow Assumption Verifier
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/smithbox-ai-controlflow-assumption-verifier/badge)](https://www.skillsdirectory.com/skills/smithbox-ai-controlflow-assumption-verifier)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: controlflow-assumption-verifier
description: "Use when a saved plan may contain hidden assumptions disguised as facts, especially for medium or large tasks, unresolved high-risk plans, or plans that mention files, APIs, versions, integrations, tests, or concurrency patterns that should be verified against the repository before execution."
---

# ControlFlow Assumption Verifier

## Overview

Hunt mirages in a saved plan. This skill adapts ControlFlow's AssumptionVerifier to Codex by checking whether the plan invents files, APIs, dependencies, conventions, or execution assumptions that the repository does not support.

## Workflow

1. Read the saved plan artifact first.
2. For each important claim in the plan:
   - identify the claim
   - decide whether it is repository-verifiable
   - verify it against files, paths, symbols, configs, or tests
   - classify it as `VERIFIED`, `UNVERIFIED`, or `MIRAGE`
3. Use [references/mirage-patterns.md](references/mirage-patterns.md) as the review grid.
4. Focus especially on:
   - phantom paths
   - phantom APIs
   - missing dependencies
   - pattern mismatches
   - missing error paths
   - missing migrations
   - missing security boundaries
5. Save the verdict to `plans/artifacts/<task-slug>/assumption-verifier.md` using `../../templates/assumption-verifier-report-template.md`.
6. Return a structured text report with:
   - `COMPLETE` or `ABSTAIN`
   - blocking and minor mirages
   - a dimensional scorecard
   - a recommendation
7. If the plan requires meaningful redesign, classify it as:
   - `fixable`
   - `needs_replan`
   - `escalate`

## Output Shape

- **Status**
- **Mirages Found**
- **Dimensional Scores**
- **Summary**
- **Failure Classification** when blocking issues exist

## Common Mistakes

- Treating every unknown as a mirage instead of separating `UNVERIFIED` from `MIRAGE`.
- Checking only presence claims and skipping missing requirements or edge cases.
- Reviewing the plan abstractly without actually opening the referenced files or paths.

## References

- `references/mirage-patterns.md`
- `../../templates/assumption-verifier-report-template.md`

Files in this skill

  • SKILL.md2.2 KB
  • references/mirage-patterns.md1.7 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.