verification-protocol

Solid

Independent verification of task completion - eliminates self-attestation

AI & Automation 335 stars 29 forks Updated today

Install

View on GitHub

Quality Score: 85/100

Stars 20%
84
Recency 20%
100
Frontmatter 20%
70
Documentation 15%
100
Issue Health 10%
80
License 10%
0
Description 5%
100

Skill Content

# Verification Protocol - Elimination of Self-Attestation ## Purpose **The Problem**: Agents were verifying their own work and always returning `success: true` by default. **The Solution**: Independent verification by a DIFFERENT agent that does NOT trust the original agent's claims. **The Rule**: `verified=true` ONLY when EVIDENCE proves all completion criteria are met. --- ## Core Principle ``` NEVER verify your own work. ALWAYS verify with independent evidence. ASSUME claims are false until proven true. Block completion without proof. ``` --- ## Verification Protocol ### Step 1: Task Completion Claim Agent claims task is complete and provides: ```json { "task_id": "task-123", "claimed_outputs": ["/path/to/file.ts", "/path/to/test.ts"], "completion_criteria": [ "file_exists:/path/to/file.ts", "no_placeholders:/path/to/file.ts", "typescript_compiles:/path/to/file.ts", "lint_passes:/path/to/file.ts", "tests_pass:/path/to/test.ts" ] } ``` ### Step 2: Independent Verification Requested Orchestrator sends to Independent Verifier Agent (different agent). ### Step 3: Verification Execution Independent Verifier checks EVERY criterion with actual evidence: ``` file_exists → fs.stat(path) && size > 0 Proof: /path/to/file.ts, 1,247 bytes, modified 2025-12-02T14:30:00Z no_placeholders → Scan for TODO, TBD, FIXME, [INSERT] Proof: 0 placeholders found typescript_compiles → npx tsc --noEmit [file] Proof: Compilation successful, 0 errors li...

Details

Author
aiskillstore
Repository
aiskillstore/marketplace
Created
5 months ago
Last Updated
today
Language
Python
License
None

Similar Skills

Semantically similar based on skill content — not just same category

AI & Automation Solid

verification-before-completion

Use when about to claim work is complete, fixed, or passing, before committing or creating PRs. Evidence before assertions.

1,160 Updated today
a5c-ai
AI & Automation Featured

verification-before-completion

Claiming work is complete without verification is dishonesty, not efficiency. Use when ANY variation of success/completion claims, ANY expression of satisfaction, or ANY positive statement about work state.

39,350 Updated today
sickn33
AI & Automation Listed

verification-agent

Design or operate adversarial verification agents that independently validate implementation work with command-backed evidence, strict read-only boundaries, and explicit PASS FAIL PARTIAL verdicts. Use when Codex needs a post-implementation verifier, QA subagent, or evidence-first acceptance gate for non-trivial changes.

1 Updated today
Beidou1507
AI & Automation Solid

verification-before-completion

Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always

2,210 Updated 1 weeks ago
foryourhealth111-pixel
AI & Automation Listed

verification-before-completion

Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always

83 Updated 3 days ago
denolfe