v0.2.1 — Dialectic Stage 2 parse tolerance

A patch release that hardens the dialectic against incomplete LLM output. Existing v0.2.0 installs will see an update-available prompt on next recall.

What's fixed

The dialectic's Stage 2 dialogue and synthesis steps were silently dropping candidates when Haiku occasionally omitted required fields (evidence on a challenger turn, action_payload on synthesis). Tolerant deserialization + the
existing validation gate now record the attempt properly, apply cooldown normally, and surface incomplete LLM output as validation_failed audit rows you can investigate, instead of having them disappear into errors_summary as parse
errors.

Impact

• No memory corruption from the prior bug — failed candidates were dropped, not mis-reframed.
• No user-facing MCP tool was affected (recall, remember, reframe, etc. all worked correctly throughout).
• This release restores Stage 2 to recording every dialectic attempt, even ones the LLM produced badly.

How to update

```bash
git pull
wrangler deploy
```

No migration, no schema changes, no config changes.

Not in this release

A separate issue where the synthesizer reaches a reframe verdict without producing the new content — also caught by the validation gate, but a prompt-design fix rather than a code fix. Tracked for the next prompt-tuning pass.

v0.2.0

What's Changed

• feat(memoria): MMR recall + REM consolidator Worker cron (CLA-87) by @JuzzyDee in #1
• Feat/cla 84 cloudflare by @JuzzyDee in #2
• feat(rem): swap REM consolidator to OAuth Bearer auth (Extra Usage credit) by @JuzzyDee in #3
• feat(rem): two-tier cluster cache — stop re-judging stable clusters (CLA-94) by @JuzzyDee in #4
• fix(oauth): consent-page hardening — XSS, redirect_uri allowlist, CSP (CLA-91) by @JuzzyDee in #5
• feat(oauth): consent-page polish — scope list + bigger button by @JuzzyDee in #6
• feat(dialectic): Stage 1 — schema + plumbing + single-agent Haiku judgment (CLA-95) by @JuzzyDee in #7
• Potential fix for code scanning alert no. 1: Workflow does not contain permissions by @JuzzyDee in #8
• feat(setup): scripts/setup.sh — one-command deploy (CLA-96) by @JuzzyDee in #11
• chore(deps): bump rustls-webpki to 0.103.13 by @JuzzyDee in #12
• chore(deps): bump rmcp 1.2 → 1.4 + unpin minor versions by @JuzzyDee in #13
• chore(deps): lockfile patches for 4 open security advisories by @JuzzyDee in #14
  :* feat(dialectic): Stage 2 — Advocate/Challenger dialogue + Synthesizer arbitration (CLA-99) by @JuzzyDee in #15
• chore(ci): trigger workflows on dev branch by @JuzzyDee in #16
• release: v0.2.0 — Oneiro by @JuzzyDee in #30

Full Changelog: https://github.com/JuzzyDee/oneiro/commits/v0.2.0