docs(spec): SPEC-008 — full-context integrity lock (agents.lock + agents.sum) - #74
Merged
Merged
Conversation
…nts.sum) Draft spec for a whole-tree provenance + integrity layer over .agents/: agents.lock records where every artifact comes from (local / source / linked / imported), agents.sum is the go.sum-style per-file content ground truth. New lock/verify commands; verify --json as a scheduled drift sentinel closing the loop with SPEC-005 quarantine. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
🤖 AI Code Review CompleteReview performed on changed files. Check inline comments for specific findings. Severity Levels:
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
Adds
specs/SPEC-008-context-integrity-lock.md(status: Draft) — a full-tree provenance + integrity layer for.agents/. Spec-only PR: no code changes. Implementation follows against #73.Design highlights
agents.lock= provenance (where every artifact comes from:local/source:<entry>/linked:<file:path>/imported:<url>, resolved SHAs, per-artifact tree hash);agents.sum= content ground truth (one<path> sha256:<hash>line per file, sorted, LF-only). Resolution and integrity stay separately reviewable — a one-line.sumchange in a PR names the exact file that moved.sources.lock, deliberately: sources.lock is the pull pipeline's per-entry resolution record for remote entries only; agents.lock is a whole-tree snapshot with different ownership, lifecycle, and failure semantics. agents.lock references sources.lock entries rather than duplicating them, and agents.sum sums sources.yaml/sources.lock (and agents.lock itself) so tampering with any layer is caught.sync-agents lock(generate/refresh),sync-agents verify(--json,--strict,--explain <path>; exit 1 on drift). pull/update/approve/detach/add/import/remove auto-refresh affected entries once the lock exists.--strict(live checkouts drift by design, SPEC-007); unlocked file appearing = ERROR (bypassed the gate)..quarantine/,.sources/, unsharedSTATE_*.md, and origin metadata are excluded; SPEC-006 OS-scoped subdirs are always locked (content record, not routing record).verify --jsonis drift detection for the whole AI-context picture.sync-agents lock.Refs #73.
🤖 Generated with Claude Code