Repository navigation
External typed-evidence evaluator invitations — wave 1 #159
Description
Activity
@scadastrangelove — you previously performed the strongest independent static review we have received and explicitly noted that no ExecSurface commands had been run.
If you are willing, we'd value a zero-assistance execution trial on any Linux x86_64 workload you choose. Please preserve the first result before asking us for troubleshooting. Failure, verifier rejection, incomplete evidence, or NO_FIT are all useful.
Packet:
03cb5dd656d6090cd2cc5c5d40885334b06b344a
Product target:cea214244d7efe05fea2108a5f5d8ed6287d9a3f
Guide: https://github.com/AETHERXGLOBAL/execsurface/blob/03cb5dd656d6090cd2cc5c5d40885334b06b344a/docs/external/P8_A3_TYPED_EVIDENCE_EXTERNAL_TRIAL.mdThis is not a request for promotion, watchlist status change, endorsement, or a positive result.
@GuillaumeFalourd — you already merged an ExecSurface GitHub Action example into
useful-actions, so you are one of the few maintainers outside AETHER X with direct context on the current public workflow.We now have a separate unreleased typed-evidence trial. If you're interested, please choose your own Linux x86_64 workload and preserve the first result before any AETHER X troubleshooting.
Packet:
03cb5dd656d6090cd2cc5c5d40885334b06b344a
Product target:cea214244d7efe05fea2108a5f5d8ed6287d9a3f
Guide: https://github.com/AETHERXGLOBAL/execsurface/blob/03cb5dd656d6090cd2cc5c5d40885334b06b344a/docs/external/P8_A3_TYPED_EVIDENCE_EXTERNAL_TRIAL.mdNo change to
useful-actionsis requested. Negative, partial, usability/friction, and NO_FIT outcomes are welcome and participation will not be described as endorsement/adoption.@astrogilda — the Probity integration you authored exposed the interoperability gap that directly motivated this typed-evidence work.
We would value a fresh consumer-side attempt against the frozen packet, using a workload you choose and preserving the initial result before AETHER X assistance.
Packet:
03cb5dd656d6090cd2cc5c5d40885334b06b344a
Product target:cea214244d7efe05fea2108a5f5d8ed6287d9a3f
Guide: https://github.com/AETHERXGLOBAL/execsurface/blob/03cb5dd656d6090cd2cc5c5d40885334b06b344a/docs/external/P8_A3_TYPED_EVIDENCE_EXTERNAL_TRIAL.mdBecause there has already been technical interaction around ExecSurface, we will qualify any returned result conservatively and separately from fully independent adoption/validation. Negative and NO_FIT results are equally useful.
@bb1nfosec — reaching out because BHEESHMA's out-of-process
strace/ptraceruntime-supply-chain model is technically adjacent to the exact boundary we are trying to falsify.We have frozen an unreleased typed-evidence trial packet for ExecSurface and are specifically looking for reviewers who may disagree with our runtime/evidence semantics.
If you are interested, please choose any Linux x86_64 workload yourself and preserve the first result before asking us for help.
Packet:
03cb5dd656d6090cd2cc5c5d40885334b06b344a
Product target:cea214244d7efe05fea2108a5f5d8ed6287d9a3f
Guide: https://github.com/AETHERXGLOBAL/execsurface/blob/03cb5dd656d6090cd2cc5c5d40885334b06b344a/docs/external/P8_A3_TYPED_EVIDENCE_EXTERNAL_TRIAL.mdA counterexample, false-completeness finding, NO_FIT/overlap conclusion, or operational criticism is at least as valuable as a successful run. No endorsement or comparison claim is requested.
Thanks for the invitation and for clearly distinguishing outreach from validation. Our review was limited to static inspection for awesome-list curation. Given our limited maintainer resources, we’re unable to participate in execution trials.
Please keep our WATCHLIST review separate from product validation and do not describe us as participating evaluators or endorsers. We’d also appreciate no further trial-related invitations or @mentions.
Thank you for understanding, and best of luck with the project.
@varunsh-coder — reaching out because Harden-Runner operates directly in the CI runtime boundary where file/network/process observations become policy-relevant evidence.
We have frozen an unreleased ExecSurface typed-evidence trial and would value a zero-assistance first run from someone who already works on runner hardening and runtime monitoring semantics.
Please choose your own Linux x86_64 workload and preserve the first result before any AETHER X troubleshooting.
Packet:
03cb5dd656d6090cd2cc5c5d40885334b06b344a
Product target:cea214244d7efe05fea2108a5f5d8ed6287d9a3f
Guide: https://github.com/AETHERXGLOBAL/execsurface/blob/03cb5dd656d6090cd2cc5c5d40885334b06b344a/docs/external/P8_A3_TYPED_EVIDENCE_EXTERNAL_TRIAL.mdA false-positive/false-completeness finding, usability failure, overlap/NO_FIT conclusion, or architecture criticism is fully acceptable. This is not a request for endorsement, promotion, integration, or a positive result.
@andrielc — your recent Tracee work on carrying FD paths with events, capture failures, truncation, storage errors and loss boundaries is directly relevant to the evidence-completeness questions we are trying to falsify.
We have frozen an unreleased ExecSurface typed-evidence trial. If you are willing, please choose any Linux x86_64 workload yourself and preserve the first result before AETHER X assistance.
Packet:
03cb5dd656d6090cd2cc5c5d40885334b06b344a
Product target:cea214244d7efe05fea2108a5f5d8ed6287d9a3f
Guide: https://github.com/AETHERXGLOBAL/execsurface/blob/03cb5dd656d6090cd2cc5c5d40885334b06b344a/docs/external/P8_A3_TYPED_EVIDENCE_EXTERNAL_TRIAL.mdWe are especially interested in criticism of event identity, path attribution, incompleteness/loss handling, or any case where the typed envelope overstates what the observer actually established. FAIL, PARTIAL and NO_FIT are first-class outcomes; no endorsement or comparison claim is requested.
@kkourt — reaching out because Tetragon's eBPF runtime-observability work sits very close to the proposition/evidence boundary ExecSurface is attempting to make explicit.
We have frozen an unreleased typed-evidence trial packet and are looking for reviewers with kernel/runtime-observation experience who may find a false-completeness or authority mistake.
If interested, please choose your own Linux x86_64 workload and preserve the first result before any AETHER X troubleshooting.
Packet:
03cb5dd656d6090cd2cc5c5d40885334b06b344a
Product target:cea214244d7efe05fea2108a5f5d8ed6287d9a3f
Guide: https://github.com/AETHERXGLOBAL/execsurface/blob/03cb5dd656d6090cd2cc5c5d40885334b06b344a/docs/external/P8_A3_TYPED_EVIDENCE_EXTERNAL_TRIAL.mdA counterexample, observer-authority criticism, incomplete result, operational failure, or NO_FIT conclusion is as useful as a successful run. We are not asking for endorsement, adoption, or a favorable comparison with Tetragon.
@adityasaky — reaching out because your work on the in-toto Attestation Framework, including review of runtime-trace semantics, is directly relevant to the distinction we are testing between an execution observation and the claims a typed evidence envelope is allowed to make.
We have frozen an unreleased ExecSurface typed-evidence trial. If you are willing, please choose any Linux x86_64 workload and preserve the initial result before asking us for troubleshooting.
Packet:
03cb5dd656d6090cd2cc5c5d40885334b06b344a
Product target:cea214244d7efe05fea2108a5f5d8ed6287d9a3f
Guide: https://github.com/AETHERXGLOBAL/execsurface/blob/03cb5dd656d6090cd2cc5c5d40885334b06b344a/docs/external/P8_A3_TYPED_EVIDENCE_EXTERNAL_TRIAL.mdWe would especially value criticism of predicate semantics, producer authority, completeness claims, consumer-verification boundaries, or any ambiguity that could let an observation be laundered into a stronger claim. FAIL, PARTIAL and NO_FIT outcomes are welcome. This is not a request for endorsement, integration, or a positive result.
Status
OPEN —
OUTBOUND_EVALUATOR_INVITATION_ONLYThis issue tracks targeted invitations for the experimental typed-evidence trial packet.
It is not evidence of validation, adoption, endorsement, reproduction, or external use.
Frozen references:
03cb5dd656d6090cd2cc5c5d40885334b06b344acea214244d7efe05fea2108a5f5d8ed6287d9a3fSelection rule
Invite only people with a concrete technical reason:
No mass mentions, no request for praise, no request for stars, and no assumption that participation is independent evidence.
Qualification boundary
Every returned result will be qualified separately under #114 / #140.
Negative, partial, unsupported and NO_FIT outcomes are first-class results.
Outreach itself remains non-qualifying.
Wave 1 roster — frozen 2026-10-06
@scadastrangelove@GuillaumeFalourd@astrogilda@bb1nfosec@varunsh-coder@andrielc@kkourt@adityasakyWave 1 is now closed to new outbound invitations unless an existing slot is explicitly retired for a documented reason. No replacement is needed for a polite refusal merely to preserve a numeric target.
Campaign success criteria
The campaign is successful only if evidence reaches all of the following:
The target is evidence quality, not response volume.