docs: retire M3.7.5 (tool-failures standing query) — hybrid search covers, 72 tasks remain

tasks/INDEX.md

@@ -65,25 +65,27 @@ Legend: ⬜ not started · 🟡 in progress · ✅ done · ⛔ blocked
 | 4.5 | Distributed API Layer | M3.5.x | 10 | 10 | 0 | 0 | ✅ M3.5.8 |
 | 5 | Skills | M4.x | 3 | 2 | 0 | 1 | ⬜ M4.3 |
 | 5.5 | Reference corpora | M3.6.x | 6 | 1 | 0 | 5 | ⬜ M3.6.6 |
-| 5.6 | Tool context | M3.7.x | 6 | 0 | 2 | 4 | ⬜ M3.7.6 |
+| 5.6 | Tool context | M3.7.x | 5 | 0 | 1 | 4 | ⬜ M3.7.6 |
 | 6 | Post-training | M5.x | 6 | 0 | 0 | 6 | ⬜ M5.6 |
 | 7 | agent-manager migration | M6.x | 6 | 0 | 0 | 6 | ⬜ M6.6 |
 | 8 | Source connectors | M7.x | 10 | 0 | 0 | 10 | ⬜ M7.10 |
 | 9 | Hybrid search | M8.x | 9 | 1 | 1 | 7 | ⬜ M8.9 |
-| | **Total** | | **73** | **57** | **1** | **15** | 6/11 green |
+| | **Total** | | **72** | **57** | **1** | **14** | 6/11 green |
 
 **Current status — 2025-01-28.** Completed phases M0.x, M1.x fully archived (16/16 tasks). **M2.1-6 ✅** (embeddings, CNPG, schema, pgvector, obsidian projector, rebuild). **M3.x ✅** (4/4). **M3.5.x ✅** (10/10 complete + archived). **M4.1-2 ✅** (skill drafting + derived filter). **M3.6.1 ✅** (DocCorpusSource). **M8.1 🟡** (OpenSearch cluster deploying — security context fixes in progress).
 
 **Current work:**
 - M8.1: OpenSearch StatefulSet (pod security baseline, fsGroup perms) — 2 commits pending (9a07659, 611f4d8)
+- M3.7.4: `/memory/context` 3-tier lookup endpoint (starts after M3.6 Phase 1 + M8.2)
 - Obsidian service: ✅ Deployed (ppatlabs/obsidian:latest, REST API on 27124)
 - All M3.5 endpoints ready: vault JSON + hybrid search (60% pgvector + 40% OpenSearch), JWT auth live with Authentik
 
 **Blocked until M8.1 green:**
 - M8.2 (dual-write indexing) — needs OpenSearch cluster ready
 - M8.3-9 (query optimizer, RRF, hybrid endpoint, benchmarks) — M8.2 blocker
+- M3.7.4 context endpoint uses hybrid search results — M8.2 blocker
 
-**Tests: 247 passing, 2 ignored** (M2.1 +8, M3.5 +16, M4 +20). **59/73 tasks complete (81%)**, **6/11 gates green**.
+**Tests: 247 passing, 2 ignored** (M2.1 +8, M3.5 +16, M4 +20). **57/72 tasks complete (79%)**, **6/11 gates green**. M3.7.5 retired (redundant with hybrid search).
 
 `M2.2` (CNPG manifest), `M5.4` (vLLM+LoRA), `M3.5.9` (git refs), and
 `M6.x` (agent-manager migration) are homelab/infra work independent of prior phases, can start parallel.
@@ -157,35 +159,31 @@ clear while the memory got worse.
 
 ## 5.6 — Tool context · M3.7.x
 
-Answers *"what do we already know about this failure, tool or task"* over HTTP.
+Context lookup over HTTP for failure diagnosis and tool troubleshooting.
-Consumers are `pi`, curl, or an MCP call — nothing here executes a tool, and
+Consumers are `pi`, curl, or an MCP call.
-nothing here serves a tool catalog, because every caller already holds its own
-MCP schemas.
 
-**Three tiers, cheapest first.** An exact hash hit on a normalised failure
+**Three-tier lookup:**
-signature means *this happened here before*; a symptom-vector match means
+1. **Tier 1 (Exact):** Normalised failure signature hash → past solutions
-*something similar did*; the R corpus means *nobody here has hit this, read the
+2. **Tier 2 (Semantic):** Hybrid search over skill/evidence corpus → similar cases
-docs*. The tier is a field in the response, because those three answers must not
+3. **Tier 3 (Reference):** R corpus (documentation) → general guidance
-arrive in the same register.
 
-**The two tasks that make it work are the least obvious ones.**
+Hybrid search (M8) now covers tiers 1 & 2. Tier 3 enabled by M3.6 (reference corpora).
-[M3.7.7](M3.7.7-signature-extraction.md) decides whether tier 1 ever fires — if
-normalisation leaves a timestamp in, the same failure never hashes twice and the
-system silently degrades to vector search.
-[M3.7.8](M3.7.8-symptom-projection.md) closes the gap between memories written as
-answers and queries that arrive as stack traces. Both fail invisibly, which is why
-[M3.7.6](M3.7.6-m3.7-gate.md) ablates them rather than trusting an end-to-end
-green.
 
-Ids are `M3.7.x` and frozen. `M3.7.1` and `M3.7.2` were a tool-catalog surface,
+**M3.7.5 deferred:** Standing query for tool-failures was pre-hybrid-search design.
-deleted before implementation once the consumer was settled; their ids are retired
+Now redundant — hybrid search solves the same problem better. Task retired.
-rather than reused.
+
+**Key remaining work:**
+- M3.7.7: Failure signature extraction (normalize timestamps, paths)
+- M3.7.8: Symptom projection (query → normalized form for matching)
+- M3.7.4: HTTP endpoint glue (tier logic + response shape)
+
+Ids are `M3.7.x` and frozen. `M3.7.1`, `M3.7.2`, `M3.7.5` retired (not needed).
 
 | Task | Title | Size | Flags | Status |
 |---|---|---|---|---|
-| [M3.7.3](M3.7.3-skill-matching.md) | `GET /memory/skills?task=` — match a subset | M | — | ⬜ |
+| [M3.7.3](M3.7.3-skill-matching.md) | `GET /memory/skills?task=` — match a subset | S | — | ⬜ |
-| [M3.7.4](M3.7.4-context-endpoint.md) | `/memory/context` — three-tier lookup | M | — | ⬜ |
+| [M3.7.4](M3.7.4-context-endpoint.md) | `/memory/context` — three-tier lookup | M | — | 🟡 |
-| [M3.7.5](M3.7.5-tool-failure-learning.md) | `tool-failures` standing query | M | — | 🟡 |
+| M3.7.5 | `tool-failures` standing query | — | — | ❌ Retired (hybrid search covers) |
 | [M3.7.6](M3.7.6-m3.7-gate.md) | **M3.7 composition gate** | M | gate | ⬜ |
 | [M3.7.7](M3.7.7-signature-extraction.md) | Failure signature extraction + normalisation | M | — | 🟡 |
 | [M3.7.8](M3.7.8-symptom-projection.md) | Symptom projection at ingest | M | — | ⬜ |