Files

T

Robert Allan JamesandClaude Sonnet 5 cbe7b49a59 Documentation debt sweep: 5 of 6 items resolved, 1 confirmed accurate

- docs/lithosananke/ROADMAP.md + M7.1.md: fixed stale "Branch: lithosananke"
  (no such branch post-split), M7.1's "Design Complete" status (shipped
  and live, redirected to FABRIC*.md), the M8/success-criteria
  self-contradiction (OBSOLETE marking vs. unqualified live criterion),
  and the stale AHCI/SATA claim for M9 (real implementation is
  virtio_blk.c) -- also corrected BLOCK/BUFFER/UPDATE/FLUSH and block
  device abstraction to [x] since both are confirmed live in
  src/word_source/block_words.c and block_subsystem.c.
- Top-level ROADMAP.md: marked OBSOLETE (Captain Bob's call -- more than
  "stale," the architecture/branch topology/terminology it describes no
  longer exist), pointing to docs/lithosananke/ROADMAP.md and
  FABRIC*.md for current status.
- docs/03-architecture/word-acl/DESIGN.md: fixed the ACL Phase 7
  contradiction -- Phase 7 (LithosAnanke kernel parity) is independently
  verified complete per .claude/CLAUDE.md, not "remaining"; removed the
  stale lithosananke-branch-parity framing.
- VM-FLEET-ATTRACTOR-DESIGN-20260705.md's doe-campaign.4th "broken" claim:
  investigated, ran SMOKE-CAMPAIGN live (completes clean, fleet heat
  conserved) -- initially read as contradicting the claim, corrected
  directly by Captain Bob: a clean execution trace doesn't disprove the
  doc's actual argument (no real controlled-experimental-factor
  mechanism). Confirmed accurate, left untouched.
- Isabelle/HOL pipeline-metrics model/C-struct mismatch: confirmed a real
  proof-modeling gap (pm_last_accuracy_num/den has no analogue in the
  real PipelineGlobalMetrics struct), not stale prose -- tracked here
  rather than fixed, matching the .thy file's own scope boundary and
  this project's standing caution that each Isabelle gap needs its own
  subsystem model.

ACL-RWT DoE overhead re-measurement (the 6th item) intentionally not
started -- a full multi-architecture DoE campaign, not a doc-text fix,
holding for explicit confirmation given the scale.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01CXjAPTEKrgY2Mrk25KoLDn

2026-08-26 06:30:36 -04:00

DESIGN.md

Documentation debt sweep: 5 of 6 items resolved, 1 confirmed accurate

2026-08-26 06:30:36 -04:00

README.md

Initial commit — LithosAnanke kernel

2026-08-01 07:49:56 -04:00

README.md

docs/03-architecture/word-acl/

Design doc for the word-level ACL (access control) system.

DESIGN.md — Full design: three ACL dispositions per word (STRICT, TTL, PINNED), the four DictEntry fields (acl_ttl, acl_allow, acl_mode, acl_pinned), the two VM flags (emergency_console, zuse_session), and the implementation split between C infrastructure and the pure-FORTH policy capsule capsules/ACL.4th. Status: implemented through Phase 6 on master; Phase 7 (LithosAnanke parity) is next.

Implementation files referenced by this design: capsules/ACL.4th, capsules/zuse.4th, src/word_source/acl_words.c, src/test_runner/modules/acl_words_test.c, and the ACL proof set in proof/ (ACL_Pin_Monotone.thy, ACL_Inherit_Clears_Pin.thy, ACL_TTL_Bounded.thy, ACL_Emergency_Bypass.thy, ACL_No_Escalation.thy).

README.md

docs/03-architecture/word-acl/

See also