Add first daily research brief and advance activity-core workplan.

Record 2026-07-24 delta brief from the baseline survey, document fi_brief_status
ops loading, and mark FI-WP-0002 T03/T06 complete after resolver and playbook work.
This commit is contained in:
tegwick 2026-07-24 00:17:44 +02:00
parent b5f911140b
commit cd29105756
3 changed files with 105 additions and 5 deletions

View file

@ -0,0 +1,76 @@
---
date: 2026-07-24
timezone: Europe/Berlin
author: grok
status: final
sources_checked:
- research/2026-07-24-baseline-field-survey.md
- public pricing/release trackers (directional)
- activity-core binky rhythm pattern
---
# Freedom Intelligence Daily Brief — 2026-07-24
> First daily brief. Field landscape is covered by the **baseline survey**
> (`research/2026-07-24-baseline-field-survey.md`). This file is a **delta** plus
> lab boot status — not a second full survey.
## Headline deltas
- **Lab live:** Freedom Intelligence foundation committed and pushed (`b5f9111`); workplans FI-WP-0001…0003 active.
- **Baseline research persisted** with P0/P1 open-weight collection recommendations and catalog candidates.
- **Activity-core path designed:** `fi-daily-research-brief` definition drafted (`enabled: false`); `fi_brief_status` resolver landing in activity-core.
- **No major same-day open-weight release** confirmed beyond the baseline window; public trackers still show a quiet open-source week after earlier 2026 waves (Gemma 4, GLM-5.1, DeepSeek V4-class, etc.).
- **Frontier pricing structure unchanged at order-of-magnitude:** ultra-cheap DeepSeek-class APIs vs expensive flagship output tokens — re-check provider pages before budget moves.
## Frontier & commercial (axis A)
| Item | Delta | Sources | Lab relevance |
| ---- | ----- | ------- | ------------- |
| Gemini 3.6 Flash (tracked ~2026-07-21) | Recent fast-tier closed release in public trackers | release trackers / Google product notes (verify) | Budget automation tier via llm-connect if quality fits |
| Claude Code default model churn (Sonnet 5-class reporting) | Harness defaults and promo pricing still fluid mid-2026 | Anthropic release notes / coding-agent digests | Compare closed agent quality bar; not a local collect |
*(No new flagship price sheet change verified for this brief day — treat baseline §2 as still current.)*
## Edge / local / open (axis B)
| Item | Delta | Sources | Lab relevance |
| ---- | ----- | ------- | ------------- |
| Open-source weekly feed quiet | Trackers report few/no new open releases *this* week | llm-stats / open LLM update pages (directional) | P0 spine from baseline remains valid; no catalog displacement today |
## Training & specialization (axis C)
*(none — no new homelab-method release flagged today.)*
## Harness & fleet (axis D)
| Item | Delta | Sources | Lab relevance |
| ---- | ----- | ------- | ------------- |
| FI daily rhythm design | activity-core emit → playbook → `fi_daily_brief` | this repo activity-definitions + docs | Enables weekday automation once resolver + path registered |
| Closed agent productization | Claude for Government / Code beta notes continue | Anthropic product updates | Axis D awareness only; sand-boxer remains Coulomb execution place |
## Collection candidates
No **new** candidates beyond baseline seed. Existing P0 catalog entries stand:
| Candidate | Source / revision | License | Est. size | Why reserve | Priority |
| --------- | ----------------- | ------- | --------- | ----------- | -------- |
| Qwen3-8B | HF Qwen/Qwen3-8B | Apache-2.0 (confirm) | ~516 GB | Default local + FT | high |
| Llama-3.2-3B-Instruct | HF meta-llama | Llama community | ~26 GB | Edge | high |
| BGE-M3 | HF BAAI/bge-m3 | MIT | ~2 GB | Multilingual RAG | high |
| DeepSeek-R1-Distill-Qwen-14B | HF deepseek-ai | MIT | ~928 GB | Local reason | high |
| nomic-embed-text-v1.5 | HF nomic-ai | Apache-2.0 | <1 GB | Light embed | high |
See `inventory/catalog/*__candidate.yaml`. **Blocked on backup storage pin** (FI-WP-0001-T04).
## Lab implications
1. **Operator:** pin backup path + soft quota in `docs/backup-storage-policy.md`.
2. **activity-core:** ship `fi_brief_status`, register external `activity-definitions/`, dry-run, then enable schedule (FI-WP-0002).
3. **After pin:** approve/download P0 (FI-WP-0003).
4. **Tomorrows brief:** pure external delta only — baseline + this boot day already recorded.
## Notes / caveats
- Same calendar day as baseline survey; external signal intentionally thin.
- Price and leaderboard figures remain directional until re-verified on official pages.

View file

@ -87,7 +87,23 @@ When activity-core emits `activity_task_spawn` with
5. Idempotency key: do not open a second brief file for the same `date`
Activity definition: `activity-definitions/fi-daily-research-brief.md`
Workplan: `workplans/FI-WP-0002-activity-core-daily-research.md`
Workplan: `workplans/FI-WP-0002-activity-core-daily-research.md`
Resolver: activity-core state-hub query `fi_brief_status` (implemented 2026-07-24)
### Ops: load this repos definitions
Dev / workstation worker:
```bash
export ACTIVITY_DEFINITION_DIRS=/home/worsch/freedom-intelligence
# then: make sync-activity-definitions (with ACTCORE_DB_URL + stack)
```
Production (railiance): either project
`activity-definitions/fi-daily-research-brief.md` into ConfigMap
`actcore-external-activity-definitions`, or mount a checkout under
`ACTIVITY_DEFINITION_DIRS` (`/etc/activity-core/external-definitions`). Keep
`enabled: false` until FI-WP-0002-T05 dry-run passes.
---

View file

@ -115,7 +115,7 @@ Wire into `docs/daily-brief-playbook.md` and `docs/state-hub-delivery.md`.
```task
id: FI-WP-0002-T03
status: todo
status: done
priority: high
```
@ -130,6 +130,9 @@ Mirror `binky_rhythm_status` pattern. Add unit tests with mocked hub.
**Done when:** resolver merged in activity-core and callable from definition params.
**Completed 2026-07-24:** `activity-core` `state_hub.py` `_fi_brief_status` + tests
(`test_fi_brief_status_*`). Ship with activity-core commit/push.
---
### T04 — Register external activity-definitions path
@ -172,7 +175,7 @@ priority: medium
```task
id: FI-WP-0002-T06
status: todo
status: done
priority: medium
```
@ -183,14 +186,19 @@ tasks (checkout path, Green/Blue profile, LLM via llm-connect if used). Prefer
**Done when:** `docs/daily-brief-playbook.md` § Automation consumer is enough for
the next agent session.
**Completed 2026-07-24:** playbook § Automation consumer + state-hub-delivery
contract; first manual brief posted with `fi_daily_brief` event.
---
## Acceptance (workplan-level)
- [x] Domain ActivityDefinition drafted (`enabled: false`)
- [x] Completion event contract documented
- [ ] Resolver + external path + dry-run enable (T03T05)
- [ ] Consumer path documented for harness (T06)
- [x] Resolver implemented in activity-core (T03) — ship with activity-core deploy
- [ ] External path + dry-run enable (T04T05)
- [x] Consumer path documented for harness (T06)
- [x] First manual daily brief + `fi_daily_brief` progress event (2026-07-24)
## Out of scope