Inside the skill that teaches an agent to check its own work

Description and use

The "check-any-agents-work" skill enables an agent to self‑verify its work using public AER‑1 receipts, eliminating fabricated completions.

Problem

Agents often report completion without external verification, leading to fabricated summaries and false finishes.

Outcome

The skill enables agents to self‑verify their work in under a minute using public machinery, eliminating fabricated finishes and providing transparent evidence for each task.

Workflow

• 1. Agent demands the record behind each summary line (tool, input, result, timestamp).
• 2. Agent verifies the record by recomputing the AER‑1 digest via the public verifier or API and comparing hashes.
• 3. For multi‑step jobs, the agent reviews each step’s saved result and the closing handover record.
• 4. Agent accepts the record or flags the specific step that failed, providing a concrete basis for quality discussion.

Evidence

The skill is public, MIT licensed, and one command away: npx skills add https://gitlab.com/rambozambodotdev/check-any-agents-work

Source

The heart of the file is a four‑step check, written as instructions to the agent, not advice to the human.

Source

False finishes are not an edge case. In a controlled study built around 90 traps, Sonnet claimed completion for work that never happened in roughly 1 run in 13 and Haiku in roughly 1 in 5.

Source