docs(rfd): promote bounded tool output RFD 115

docs/rfd/.priority.json

@@ -125,7 +125,7 @@
     "099",
     "104",
     "105",
-    "D24",
+    "115",
     "D34",
     "D38",
     "D46",

docs/rfd/109-in-process-jp-mcp-server.md

@@ -4,6 +4,7 @@
 - **Category**: Design
 - **Authors**: Jean Mertz <git@jeanmertz.com>
 - **Date**: 2026-09-12
+- **Extended by**: [RFD 115]
 - **Summary**: Tool execution moves into an in-process JP MCP Server that JP and
   third-party MCP clients both call over loopback Streamable HTTP.
 
@@ -473,4 +474,5 @@ shared contracts and execution behavior specified here.
 [RFD 058]: 058-typed-content-blocks-for-tool-responses.md
 [RFD 065]: 065-typed-resource-model-for-attachments.md
 [RFD 111]: 111-transitional-jp-protocol-bridge-for-mcp-tools.md
+[RFD 115]: 115-bounded-tool-output.md
 [Streamable HTTP]: https://modelcontextprotocol.io/specification/2025-11-25/basic/transports#streamable-http

docs/rfd/drafts/D70-provider-native-token-counting.md

@@ -41,7 +41,7 @@ that number is exactly what a provider counting endpoint returns.
 
 Doing nothing leaves the search manual and leaves every future consumer
 estimating.
-Two other drafts, `D24` (bounded tool output) and `D50` (automatic compaction),
+Two other drafts, `115` (bounded tool output) and `D50` (automatic compaction),
 each defer accurate counting for the same stated reason — a tokenizer is a real
 dependency — and each works around its absence with a safety margin.
 

docs/ticket/0n0nwz3-harness-design-findings-from-arxiv-2609-20804-measured-again.md

@@ -67,7 +67,7 @@ The proposals it produced are filed separately and listed at the bottom.
 | M2 recall (`recall_event` tool)                                            | absent; `--reset` plus always-preserved raw events                             | correct by design                                 |
 | M3 summarization (running summary)                                         | `SummaryPolicy`                                                                | exists, manual only                               |
 | Soft / hard thresholds (0.6 / 0.85), verbatim recent window (0.3, floor 2) | none                                                                           | RFD D50 proposes a single `trigger_ratio`         |
-| Per-result cap (24k chars)                                                 | none                                                                           | RFD D24                                           |
+| Per-result cap (24k chars)                                                 | none                                                                           | RFD 115                                           |
 | Truncation fallback                                                        | `window::truncate_to_fit`                                                      | exists; two call sites, neither on the query path |
 
 JP has every mechanism the paper tested.
@@ -117,7 +117,7 @@ It has no trigger.
 
 - **No context management on the query path.** The failure the paper measures
   most directly, and T-0de0hry is in-repo evidence it already bites.
-- **Unbounded tool output.** RFD D24 records a 1,293,623-token request against a
+- **Unbounded tool output.** RFD 115 records a 1,293,623-token request against a
   1,000,000 limit, durably persisted.
 - **No loop bound and no repetition detection.** The turn loop cycles streaming
   to executing with no cap.
@@ -172,7 +172,7 @@ Ordered by leverage per unit of cost.
 The first two attack the failure the paper measures most directly; the third is
 what makes any of them checkable.
 
-1. **T-0n0pcr0**: promote D24 and rank bounded tool output.
+1. **T-0n0pcr0**: promote 115 and rank bounded tool output.
    The cheapest change, and it sits upstream of the rest: a single oversized
    response poisons a conversation permanently, and no compaction trigger
    rescues one that has already happened.

docs/ticket/0n0pcr0-promote-d24-and-rank-bounded-tool-output-for-implementation.md

@@ -1,4 +1,4 @@
-# Promote D24 and rank bounded tool output for implementation
+# Promote 115 and rank bounded tool output for implementation
 
 - **Status**: Todo
 - **Kind**: Chore
@@ -14,10 +14,10 @@ JP places no ceiling on a tool call response.
 flushes it, so an oversized response is durable: every later turn re-sends it,
 the provider rejects the request, and the conversation needs hand-editing to
 recover.
-RFD D24 records the case that prompted it, a 1,293,623-token request against a
+RFD 115 records the case that prompted it, a 1,293,623-token request against a
 1,000,000-token limit.
 
-D24 has been a Draft since 2026-07-27 and sits unranked in the backlog.
+115 has been a Draft since 2026-07-27 and sits unranked in the backlog.
 
 ## Why now
 
@@ -37,7 +37,7 @@ new axis.
 
 ## What to do
 
-1. Promote D24 to Discussion.
+1. Promote 115 to Discussion.
 2. Rank it.
    It gates the value of everything else done about context pressure: an
    automatic compaction trigger that fires on a conversation already poisoned by