Verification Loop — Claim-to-Proof Workflow
Tie each completion claim to observed verification evidence.
$5
Works with the AI Agents you already use
Claude Code
Cursor
Codex CLI
Muse
Dots+22 moreClaude CodeCursorCodex CLIMuseDots
OpenClaw
Manus
Grok
Grok Bot+18 agentsVerification Loop — Claim-to-Proof Workflow
Example session with this skill installed
Illustrative scenario: Review whether an example bug-fix task meets each acceptance criterion.
- Read your context and instructions
- Compiled the verification loop
ILLUSTRATIVE FORMAT ONLY — no commands, model calls or application runtime were executed. This is a proposed output structure, not a verified session.
Proposed acceptance record
Claim: the reported bug is fixed.
Verification action: identify a relevant reproducible test or authorized manual check.
Observed evidence: leave empty until the action actually runs and produces inspectable output.
Status: Not verified in this illustrative example.
Additional criteria: record regression coverage and any unavailable tools separately. Failed checks remain Failed; blocked checks remain Could not verify. A plausible implementation is not proof of completion.
Evidence boundary
Populate this record from observed files, tool output or user-provided facts. Keep unavailable evidence and unresolved questions explicit. Do not invent successful tests, generated artifacts, source citations or performance improvements.
Next steps
Set up the documented dependencies in a compatible agent environment, confirm the authorized scope, and perform a small real task. Review the actual output before relying on it. Agensi runtime integration is untested; no hosted service, credentials or API credits are included.
Connects securely to your tools. The creator never sees your data.
About this skill
Tie each completion claim to observed verification evidence.
Includes 1 original skill(s), one root routing entry, original source/runtime helpers where applicable, license notices, installation guidance and a source-integrity manifest.
Free upstream source is publicly available: https://github.com/MiniMax-AI/MiniMax-Code-Plugins/tree/6481e4ae1e0375a4ddb3d5e9b29845354ab1cead/plugins/Hylouis233/verification-loop. License: Apache-2.0. This packaged distribution does not grant exclusivity or remove the upstream open-source rights.
Dependencies: A coding-agent host with appropriate authorized test/read tools. No bundled runtime service.
Limitations: Verification may require project-specific tools. No guarantee of software correctness; remote checks remain subject to explicit authorization in the skill.
Agensi runtime integration has not been tested. Packaging received static integrity and secret-pattern review only; no application runtime or external model was executed. No API keys, API credits, model subscriptions, hosted service or third-party data access are included.
Price: US$5 one-time purchase for the complete package, not a subscription.
What's included (2 skills in this plugin)
verification-loop
Tie each completion claim to observed verification evidence.
verification-loop
Enforce a claim-to-proof workflow before declaring work complete. Use this Skill when a task involves implementation, debugging, review, or any other work that could be claimed complete without concrete evidence.
How to use
Skills on Agensi run inside your AI agent. Copy the prompt, paste it into your agent, and it connects to Agensi and installs this skill for you. If a payment is needed, your agent shows it to you first.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill passes our trust checks below.
Security scanned
Passed our 8-point scan before listing
Fresh listing
Recently published to Agensi
30-day refund
Not a fit? Get your money back
Trust & safety
Security scanned
Verified clean today
- Passed all security checks, Safe to install