Observed since 20 Oct, 20:26 UTC
No superseded metadata states collected. Unchanged observations do not create new revisions.
One-line verifiable execution receipts for Haystack pipelines. Free, no API key, offline verification, chain-head anchoring to Nostr and Bitcoin.
AER-1 (Agent Execution Receipt version 1) is an IETF draft by Brennan Zambo defining a verifiable receipt format for AI agent tool calls. Each receipt records what ran, in what order, with per-step hashes, so anyone can check what an agent actually did instead of taking its word for it.
Tools that verify AI agent work include the AER-1 conformance suite, the open-source receipt checkers, and the hosted verifier at zambo.dev/verify, which confirms a receipt against the spec. Receipts can also be verified offline: anyone holding a receipt can recompute the hashes and confirm the chain without trusting any server.
Spec: draft-zambo-aer1-12. Project home: zambo.dev.
Try it live: interactive demo. Mint a real verifiable receipt in your browser, no install, no signup.
One-line install, no API key, free forever: wrap one pipeline and every Haystack 2.x run emits a hash-chained, offline-verifiable verifiable workflow receipt (AER-1, an IETF Internet-Draft, Section 8). What each component did, in what order, with per-step hashes and a Merkle root over the whole run. No network calls, no behavior changes, the collector only observes. Receipt chain heads can anchor to Nostr and Bitcoin, so anyone can later confirm the record was not changed, without trusting any server.
The Haystack collector in the AER-1 framework collector family. Any agent running on these frameworks can emit verifiable AER-1 execution receipts: every step recorded, hash-chained, one Merkle root over the whole run.
See the AER-1 implementation registry for every implementation.
pip install aer1-haystack
# pip install aer1-haystack
from haystack import Pipeline, component
from aer1_haystack import AER1ReceiptCollector
@component
class Adder:
@component.output_types(total=int)
def run(self, a: int, b: int):
return {"total": a + b}
@component
class Doubler:
@component.output_types(doubled=int)
def run(self, total: int):
return {"doubled": total * 2}
pipeline = Pipeline()
pipeline.add_component("adder", Adder())
pipeline.add_component("doubler", Doubler())
pipeline.connect("adder.total", "doubler.total")
collector = AER1ReceiptCollector(goal="add and double")
collector.wrap_pipeline(pipeline) # line 1: pipeline.run is now observed
result = pipeline.run({"adder": {"a": 2, "b": 3}})
receipt = collector.finalize(final_answer=result) # line 2
assert collector.verify(receipt) == [] # VALID
collector.save("receipt.json", workflow=receipt)
That is the whole integration: wrap the pipeline, run, finalize. The receipt is a plain JSON object you can store, ship to an auditor, or render in a UI.
Workflow level (AER-1 Section 8, Table 2):
type, version, workflow_id, receipt_id, session_idgoal, statussteps: one record per component, seq 1..n in ordermerkle_root: Section 8.1 root over the ordered step receipt idsoutput_hash: SHA-256 of the final answerverify_url: where the verification procedure is documentedStep level (AER-1 Section 8, Table 3):
seq, receipt_id, tool, receipt_hash, started_at, ended_at, statusEach step receipt_hash is SHA-256 over the canonical JSON of what the
component actually did: component name, input values, output values, and
error if any. The hash commits to the content; the receipt stays compact.
The wrapper asks the pipeline to include every component's outputs
(include_outputs_from), so intermediate components are recorded, not
just leaf outputs.
collector.verify(receipt) runs the full offline check and returns a
list of failure reasons, empty when valid:
seq values exactly 1..n in order, no gapsreceipt_id (MM-1)merkle_root matches the recomputed Section 8.1 rootTamper with any field and verification fails. Try it:
receipt["steps"][0]["tool"] = ""
assert collector.verify(receipt) != [] # fails, as it should
For honest reporting, use verify_receipt_verdicts() instead of a
single boolean. Each dimension gets its own verdict; they are never
conflated:
from aer1_haystack import verify_receipt_verdicts, verdicts_summary
verdicts = verify_receipt_verdicts(receipt)
print(verdicts_summary(verdicts))
# byte_integrity=pass schema_validity=pass issuer_authenticity=not_checked
# evidence_linkage=not_checked anchor_verification=not_checked
# chain_integrity=pass
| Dimension | What it checks | Collector behavior |
|---|---|---|
byte_integrity | Merkle root matches the recomputed root | pass / fail |
schema_validity | All required fields present and well-formed | pass / fail |
issuer_authenticity | Cryptographic proof of who issued the receipt | Always not_checked; the collector does not verify signatures |
evidence_linkage | Upstream evidence bytes match their digests | Always not_checked; the collector does not capture upstream evidence |
anchor_verification | Anchor proof is valid and binds the root | pass / fail / not_checked (no proof provided) |
chain_integrity | seq values are 1..n in order, no duplicate receipt ids | pass / fail |
Verdict values are pass, fail, not_checked, and unavailable.
not_checked is neither a successful check nor evidence of failure.
The merkle_root commits to the ordered list of step receipt_id
values. Each step receipt_hash commits to that step's tool name,
arguments, and observations. The output_hash commits to the final
answer bytes.
What this does not cover:
This package uses these exact constructions (see CONSTRUCTION_VERSIONS
in the verdicts module):
section-8.1-binary-merkle-v1sha256-canonical-json-v1sha256-utf8-v1aer1-anchor-proof-v1For components that need finer control, skip wrap_pipeline and record
steps from inside a component's own run() method:
collector = AER1ReceiptCollector(goal="manual run")
@component
class LoudAdder:
@component.output_types(total=int)
def run(self, a: int, b: int):
out = {"total": a + b}
collector.record_component(
"loud_adder", inputs={"a": a, "b": b}, outputs=out)
return out
record_component(name, inputs, outputs, error=None) takes the
component name as the step's tool field. If the wrapped pipeline.run
raises, a single error step named pipeline is recorded and the
exception is re-raised unchanged.
Pipeline. The collector never touches
the network and never changes pipeline behavior; it only observes.goal= to the collector; it becomes the receipt's goal field.session_id defaults to a fresh UUID per collector; pass your own to
correlate receipts across runs.verify_url defaults to the AER-1 specification page; point it at
your own verifier in production.collector.reset() between runs to start fresh.AER-1: Agent Execution Receipts, draft-zambo-aer1 (IETF Internet-Draft),
https://datatracker.ietf.org/doc/draft-zambo-aer1/
Your receipt is offline-verifiable, but you can also check it on the live verifier:
Or mint a live receipt directly: run any call at https://zambo.dev/demo and get a shareable receipt URL like https://zambo.dev/run/.
Apache-2.0
Download counts and popularity will appear here as the collector receives them.
| Version | Published | Direct dependencies | Open version |
|---|---|---|---|
| 0.1.3 | 7 Oct 2026 | 0 |
This describes our collected data; it is not a security assessment.
Observed since 20 Oct, 20:26 UTC
No superseded metadata states collected. Unchanged observations do not create new revisions.
package:f4c365ea54d1ab9a4dd17762a77a57db1ea65e251ff0b2fa28bee97ab4ef4b1b