From e789f6f7ea23af9b277b1ebf57a2f25ce08ebaf6 Mon Sep 17 00:00:00 2001 From: Omar Sobh Date: Tue, 22 Sep 2026 15:50:29 -0500 Subject: [PATCH] docs: continuous_improvement reaches its subject; the brief, not the role prompt, aims it v2 got the record delivered but never read it (0/50 calls; role prompts are inert on missions and the reused v1 brief pointed at brains). v3, with a brief aimed at PROJECT-MEMORY.md, read it, reported 4 MET / 0 UNMET and proposed nothing, and was failed by the judge for summarising rather than quoting. Co-Authored-By: Claude Opus 5.5 (1M context) Claude-Session: https://claude.ai/code/session_01WZb5A2kfVfjpdwSochkuHz --- docs/TEMPLATE-MATURITY.md | 30 +++++++++++++++++++++++++++++- 1 file changed, 29 insertions(+), 1 deletion(-) diff --git a/docs/TEMPLATE-MATURITY.md b/docs/TEMPLATE-MATURITY.md index fbc0946..5501874 100644 --- a/docs/TEMPLATE-MATURITY.md +++ b/docs/TEMPLATE-MATURITY.md @@ -65,7 +65,7 @@ holding (it was 55 of 85 dangling once). | `codebase_research` | code_archeologist architecture_mapper flow_tracer vault_scribe | research_readonly | 11 | **yes** — first run 2026-09-22 | | `papers_research` | domain_scout paper_reader library_curator | research_web_readonly | 8 | **yes** — first run 2026-09-22 | | `insight_research` | implementation_tracker novelty_hunter publication_drafter | research_readonly | 8 | no | -| `continuous_improvement` | brain_inspector improvement_proposer improvement_evaluator | research_readonly | 7 | **runs, cannot reach its subject** | +| `continuous_improvement` | brain_inspector improvement_proposer improvement_evaluator | research_readonly | 7 | **reaches its subject; right answer, judge-failed on quoting** | **Six of twelve are evidenced.** `backend` was exercised on 2026-09-22, the first time anything had been staffed from it: all five roles @@ -122,6 +122,34 @@ template assumes long-lived agents that accumulate a record; the platform makes disposable ones. That is a design decision to resolve, not a bug to patch. +**Retargeted (2026-09-22).** The decision taken: audit the *repo* brain, +which is where missions actually write what they learn. The server now +exports it as `/mission/memory/PROJECT-MEMORY.md`, one line per judged +phase, and installs it into every mission that has a repository. + +Two runs followed, and they separate the channel from the steering: + +- **v2 (01a0cad8) — channel works, steering does not.** The 2.4 KB record + was installed. The agent made 50 tool calls, 0 of them touched it, 13 + went to the roster, and the judge passed it. The brief and `done_when` + were reused from v1 ("read each agent's brain"), and the template's + retargeted role prompts are **inert on mission turns** (see + `prompt-injection-two-paths`). Rewriting `system_prompt` cannot aim a + mission; only the brief can, and nothing supplies one for this team. +- **v3 (01a0cade) — right answer, failed by the judge.** With a brief and + `done_when` pointing at the record, 3 of 7 tool calls read it. The audit + states the record exactly (4 lines, MET 4 / UNMET 0), finds no pattern, + and files **zero** proposals ("a single UNMET line would be an incident; + this record has zero"), the correct result on an all-MET record. The + judge failed it, correctly by the letter: the condition required each + finding to *quote* the record lines, and the audit summarised them in a + table instead. `max_iterations` was 1, so there was no retry. + +So the subject is now reachable and the reasoning holds up. The unproven +part is the case this team exists for, a record *with* failures. That +needs a repository whose history contains UNMET verdicts, and a default +brief so no caller has to write one. Still not counted as evidenced. + **Two of the remaining six are still unevidenced**, and the other four are blocked on a target stack. The other nine are well-formed scaffolding: roles, prompts, brain seeds and resolving skills, and no run behind any of