Skip to content

Release pf-core 0.7.0 reward-binding (multi-commit) - #1

Merged
fraware merged 18 commits into
mainfrom
release/pf-core-0.7.0-reward-binding
Jul 24, 2026
Merged

Release pf-core 0.7.0 reward-binding (multi-commit)#1
fraware merged 18 commits into
mainfrom
release/pf-core-0.7.0-reward-binding

Conversation

@fraware

@fraware fraware commented Jul 24, 2026

Copy link
Copy Markdown
Member

Summary

  • Rebuilds the pf-core 0.7.0 reward-binding release as 18 focused commits (PF-VA-00 through VA-08) with the same tree as the prior monolithic release on main.
  • Covers ADR/baseline docs, PCS schemas and adapter, validator/CLI, Lean RewardBinding/certificate/soundness/replay, adversarial and companion fixtures, and VERSION/CHANGELOG closeout.
  • Merge with a merge commit (not squash) so the granular history lands on main.

Test plan

  • Confirm PR tree matches prior release commit 0330172
  • Spot-check commit sequence follows PF-VA boundaries
  • After merge, verify main contains all 18 commits via merge commit

fraware added 18 commits July 24, 2026 10:37
Record ownership boundaries for Lean model, schemas, CLI, and PCS adapter so release work has a single decision record.
Capture the starting scope, deliverables, and acceptance checklist that later VA steps build on.
Align product claims with what the Lean model and validator actually certify versus what remains out of scope.
Document A11-A14-facing assumptions and the release gates needed to ship pf-core 0.7.0 reward-binding.
Pin authority, environment, verifier profile, verification result, and reward evidence envelope schemas for adapter consumers.
Normalize PCS artifacts into trusted reward-binding inputs and ship a known-good fixture set for adapter and validator checks.
Define the trusted input shape validators consume, with valid and invalid examples plus schema-map documentation.
Evaluate binding, integrity, issuer, and profile checks in Python and emit structured decisions covered by unit tests.
Expose reward-binding validation through the pf-core CLI so operators can run the same predicates as the library API.
Formalize reward-binding structures and extend assumptions so later certificate and soundness theorems have a stable base.
Prove certificate well-formedness and soundness relative to the reward-binding model, including A11-A14-facing results.
Encode replayable golden cases that lock expected accept/reject behavior for the formal certificate path.
Serialize accepted reward-binding decisions into certificates that align with the Lean certificate shape for offline replay.
Map Lean theorems to runtime checks and clarify adversarial assumptions for reward-binding certificates.
Cover revoked issuers, trace substitution, profile conflicts, staleness, and related reject paths for adapter regression.
Provide handoff fixtures for LabTrust reward-binding wiring and OVK verifier-profile pinning without coupling release to those repos.
Spell out follow-on PR scope so parent repos can land schema and profile pins after pf-core 0.7.0.
Record the reward-binding release surface, security notes, and final inventory for PF-VA-08 closeout.
@fraware
fraware merged commit b853bea into main Jul 24, 2026
2 of 6 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant