osobh/clawmates
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

docs: frontend team evidenced — 8 of 12, judged by the Kimi fallback
deploy / test (push) Successful in 5m6s
Details
deploy / build (push) Canceled after 51s
Details

Browse Source 
Mission 01a0ce4c: offline npm install, GLM 1310, fallback to kimi-for-coding,
suite re-run by the judge, MET and independent on the first pass.

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>
This commit is contained in:
Omar Sobh
2026-09-23 07:49:10 -05:00
co-authored by Claude Opus 5.5
parent abfba832e1
commit dd335ccfe6
1 changed files with 8 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/TEMPLATE-MATURITY.md
+8 -1
View File
@@ -59,7 +59,7 @@ holding (it was 55 of 85 dangling once).
| `topic_research` | lead_researcher evidence_checker report_writer | research_web_readonly | 4 | **yes** — measured | | `topic_research` | lead_researcher evidence_checker report_writer | research_web_readonly | 4 | **yes** — measured |
| `continuous_research` | paper_reader signal_ranker script_writer | research_readonly | 11 | **yes** — live runs | | `continuous_research` | paper_reader signal_ranker script_writer | research_readonly | 11 | **yes** — live runs |
| `backend` | api_designer db_engineer coder tester committer | coding_readwrite | 20 | **yes** — first run 2026-09-22 | | `backend` | api_designer db_engineer coder tester committer | coding_readwrite | 20 | **yes** — first run 2026-09-22 |
| `frontend` | designer coder tester committer | coding_readwrite | 12 | **work verified by hand; judge cannot run npm** | | `frontend` | designer coder tester committer | coding_readwrite | 12 | **yes** — first run 2026-09-23 (judged by the Kimi fallback) |
| `mobile` | designer coder tester committer | coding_readwrite | 11 | no | | `mobile` | designer coder tester committer | coding_readwrite | 11 | no |
| `gpu` | arch_analyst kernel_author bench_engineer coder committer | coding_readwrite | 13 | no | | `gpu` | arch_analyst kernel_author bench_engineer coder committer | coding_readwrite | 13 | no |
| `threejs` | scene_designer coder shader_author perf_engineer committer | coding_readwrite | 13 | no | | `threejs` | scene_designer coder shader_author perf_engineer committer | coding_readwrite | 13 | no |
@@ -262,6 +262,13 @@ its pass unspent, as designed. The agents' work was green again: 10 tests,
`tsc -b` clean, +221 lines delivered. Evidenced once a judge with quota `tsc -b` clean, +221 lines delivered. Evidenced once a judge with quota
passes it. passes it.
**Evidenced (mission 01a0ce4c).** Run a third time once the Kimi fallback
judge was live (abfba83). The whole chain fired in order: the harness installed
dependencies offline, GLM refused with 1310, the evaluator fell back to
`kimi-for-coding`, and Kimi re-ran the suite itself (*"All claims verified by
direct inspection and test run"*). Verdict MET, `independent = true`, on
iteration 0. **Eight of twelve.**
## One thing both runs showed: agents reach for Bash ## One thing both runs showed: agents reach for Bash
Cumulative tool calls across every mission since the policy shipped: Cumulative tool calls across every mission since the policy shipped: