governance-receipt-recovery

Recovers lost governance thread context from receipts and disk pointers.

Updated Jun 20, 2026
One-click install
npx skills add https://github.com/Noetfield-Systems/SourceA --skill governance-receipt-recovery-noetfield-systems
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: governance-receipt-recovery
Source: https://github.com/Noetfield-Systems/SourceA/tree/main/cursor-plugin/sourcea-forge-governance/skills/governance-receipt-recovery
Command: npx skills add https://github.com/Noetfield-Systems/SourceA --skill governance-receipt-recovery-noetfield-systems

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? When a conversation loses context—after summarization, a user correction, or a missed link—governance and audit work stalls because claims can no longer be traced to proof. This Skill restores continuity by searching prior receipts, registry JSON, and README paths on disk before any reply is made. ## Core Features & Use Cases - Structured recovery loop: Pause, search receipts and registry, report FOUND or NOT FOUND with exact paths, then prove claims via verification tools. - Receipt verification: Uses verify_run to load receipts by id and emit_receipt_readonly to validate schema before citing a receipt as proof. - Honest verdicts: Distinguishes PASS, FAIL, and MOCK_ONLY outcomes, and never invents receipt ids or claims enterprise federation before cloud storage is live. - Use Case: After a mid-audit context summarization, the user says "you missed the receipt." The Skill re-reads the receipt and registry, returns the exact path and verdict, and proposes the next single action. ## Quick Start Ask the assistant to recover the lost governance receipt from disk and prove the verdict before continuing the audit.

Frequently Asked Questions about governance-receipt-recovery

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I recover lost context after conversation summarization?

Re-read the prior receipts and registry JSON from disk instead of relying on the chat tail summary. The recovery loop searches indexed paths, names the exact file or receipt id found, and verifies it before replying.

How do I verify a governance receipt before citing it as proof?

Use verify_run to load a receipt by id or validate an inline witness, and emit_receipt_readonly to check the schema. Only cite the receipt as proof after one of these read-only checks passes.

What does a MOCK_ONLY verdict mean in governance audits?

MOCK_ONLY means the claim was verified against local stdio-tier evidence, not a live cloud bucket. It should be explained honestly rather than presented as enterprise-safe federation.

When should receipt recovery run before writing a new governance document?

Run it whenever the user corrects a missed item, after mid-audit context summarization, or before creating a new governance doc. Checking the existing registry row first prevents duplicate law prose and broken pointers.

What are the limitations of receipt-based context recovery?

Recovery depends on receipts and registry files actually existing on disk; it cannot reconstruct context that was never persisted. It also cannot claim cloud-tier federation until the cloud bucket is live.