Restack knowledge half-life onto current main (#569)

[davekilleen]

Linked Issue

• Restack of Proposal: declare how long an observation stays trustworthy, and measure it #569 after Re-state the time on every prompt, so a long session cannot lose the date #566 and Capture corrections when they happen, not at the end of the day #568 landed
• Linear issue: none (Track C learning stack)

What Changed

• Cherry-picked the three real Proposal: declare how long an observation stays trustworthy, and measure it #569 commits onto current origin/main (8cf7dd84, the Capture corrections when they happen, not at the end of the day #568 squash)
• Dropped the duplicated unsquashed Capture corrections when they happen, not at the end of the day #568 commits that made Proposal: declare how long an observation stays trustworthy, and measure it #569 DIRTY
• Intended behavior only: knowledge half-life seed YAML, freshness machinery, observation-recorder hook file (left unwired in settings), and generated inventory/catalog/contract updates
• Did not invent skill wiring Chris left unwired

Test Plan

• Unit/integration tests added or updated: existing core/tests/test_freshness.py carried forward
• Negative/error-path tests added or updated: unchanged from Proposal: declare how long an observation stays trustworthy, and measure it #569
• Commands run locally: git merge-tree of this branch onto origin/main is clean; compare is 3 commits ahead, 0 behind

Ralph Wiggum Loop

• I implemented the change.
• I self-reviewed for defects and edge cases.
• I requested specialist review for risky areas (testing/infra/security when relevant).
• I addressed review findings and re-ran checks.

Quality Gates

• I added/updated tests or documented why no tests are needed.
• I added a regression test for bug fixes, or this PR is not a bug fix.
• I validated failure modes / edge cases.
• I updated docs or confirmed no docs impact.
• CI checks for lint + tests + coverage are expected to pass.

Risk & Rollback

• Risk level (low/medium/high): low
• Rollback plan: close this PR; the three commits are only additive seed/machinery files

Docs Impact

• Files updated: docs/architecture/INVENTORY.md (generated), docs/examples/dex-lens-catalog-enriched-preview.json (generated)
• If none, reason: n/a

Notes

• Do not merge the dirty fork PR Proposal: declare how long an observation stays trustworthy, and measure it #569 until its head can be updated
• Same-origin copy also at knowledge-half-life (f35e24df)
• Does not touch usage PRs The feature-adoption log had a reader, a documented writer, and no way to reach it #590 / Nothing calls the adoption tool, so skills still hand-edit the log #593

  

[github-actions]

What this pull request touches

• session hooks — feeds starting sessions and carrying useful context into the next interaction.
• tests — feeds catching regressions before contributors and users encounter them.

Gates that will judge this change

• Personal-data gate: added lines must not expose real identities or personal vault content.
• Change-aware gates: source changes are checked for tests, path-contract use, documentation drift, and touched-file coverage.
• Tests and coverage: the Python, MCP, migration, hook, and script suites must remain healthy.
• Safety and quality: security, lint, distribution, path consistency, and large-vault checks still apply.

Based on 10 changed files.

[cursor]

Approved. Cursor Bugbot was not present after the first check poll, so that signal was skipped; no applicable approval policy required human review. No reviewers were assigned.

  

_{Sent by Cursor Approval Agent: Pull Request Router and Approver}