Agent coordination audit ACA-1 with signed receipt
For anyone who runs more than one agent, or one agent among strangers: declare how your agents claim work, retry, hand over, verify and stop, and get an audit against sixteen measured patterns of lost work: duplicated effort, repeated side effects, loops, amplified errors, lost context, locks nobody releases. Returns verdict, score, every open finding with the measurement behind it, its fix and the move that closes it, the first three steps and an ed25519-signed receipt. Unknown counts as open. Free preview at /v1/coordination/preview, questions at /v1/coordination/protocol.
Request
- Base URL: https://verify.astranl.com
- URL: https://verify.astranl.com/v1/agent-coordination-audit
- Auth: none declared
Query parameters
yes, partial, no or unknown. Before an agent starts a piece of work that another agent could also pick, does it take a claim that the others can see?
yes, partial, no or unknown. Do claims and locks expire by themselves unless the holder refreshes them?
yes, partial, no or unknown. Does every action with an outside effect, a payment, an order, a message, a write, carry a key that makes a repeat harmless?
yes, partial, no or unknown. Is there an explicit end state for every task, checked by the runtime, with a ceiling on steps and on repeated identical calls?
yes, partial, no or unknown. Is the result of an agent checked by something other than that agent before it is used or passed on?
yes, partial, no or unknown. When work passes between agents, does it pass as a fixed structure with the artifacts by reference, not as a retelling?
yes, partial, no or unknown. Does every delegated task carry a remaining hop count or budget that is reduced at each handoff and stops the chain at zero?
yes, partial, no or unknown. Before splitting work across agents, is there a rule that decides from the task whether more agents help at all?
yes, partial, no or unknown. Is the input of each step measured and bounded: trajectory pruned, cache hit rate watched, tool definitions loaded on demand?
yes, partial, no or unknown. On a rate limit, a timeout or a lost claim, do agents back off with a random, growing wait, and is the number of concurrent calls limited?
yes, partial, no or unknown. Before an agent writes to shared state, does it check that what it read is still current?
yes, partial, no or unknown. Does finished work come with a receipt another party can check: what was asked, what was delivered, by whom, when?
yes, partial, no or unknown. Before work or payment, is it checked that the endpoint answers, that the reward is funded and how many others are already on it?
yes, partial, no or unknown. Is what worked and what failed recorded where the next run or the next agent will read it?
yes, partial, no or unknown. Are routine actions allowed by standing policy and a sandbox, with human approval kept for the irreversible ones?
yes, partial, no or unknown. Is the same task run several times before its cost and success rate are trusted?
comma list of parallel, tools, handoff, market; default all
optional name of the audited system, echoed in the receipt
optional number of agents in it
Response
signed verdict
Changes
No recorded changes to this operation across all 1 revision of this API.