git:20260808.823116a to git:20260816.ab9dddc
11 added, 11 removed. Audit A to A.
---
name: tlamatini-exec-report-row-adder
description: Add a state-changing tool to _EXEC_REPORT_TOOLS in mcp_agent.py and the matching CSS rules so its operations appear in the chat-page Exec Report.
metadata:
openclaw:
emoji: "📋"
tlamatini:
runtime: in-process
requires_tools: ["chat_agent_file_creator","chat_agent_executer"]
requires_mcps: []
budget:
max_iterations: 6
max_seconds: 60
max_tokens: 10000
permissions:
filesystem:
read: ["Tlamatini/agent/mcp_agent.py","Tlamatini/agent/static/agent/css/agent_page.css","Tlamatini/agent/static/agent/css/agentic_control_panel.css"]
write: ["Tlamatini/agent/mcp_agent.py","Tlamatini/agent/static/agent/css/agent_page.css"]
shell:
- "python Tlamatini/manage.py test agent.tests.ExecReportCaptureTests"
network: deny
db: deny
inputs:
- { name: tool_name, type: string, required: true }
- { name: agent_key, type: string, required: true }
- { name: agent_display, type: string, required: true }
outputs:
- { name: rows_added, type: integer, required: true }
- { name: tests_pass, type: boolean, required: true }
triggers:
keywords: ["exec report","add to exec report","_EXEC_REPORT_TOOLS"]
---
<!--
═══════════════════════════════════════════════════════════════════
✦ T L A M A T I N I ✦ — "one who knows"
Created by Angela López Mendoza · @angelahack1
Developer · Architect · Creator of Tlamatini
Tlamatini Author Banner — do not remove (Angela's name is kept in every build)
═══════════════════════════════════════════════════════════════════
-->
# Exec Report row adder
> **Read this first — capture is ALREADY automatic (2026-06-07).** EVERY wrapped
> `chat_agent_*` is captured by `_resolve_exec_report_spec` with no code at all,
> observational/output and read-only agents INCLUDED. This procedure is an
> **OPTIONAL refinement**: use it only to merge a shared `agent_key` (a direct
> `@tool` + its wrapped launch), fix the display casing, or give the table a
> CSS-matched caption gradient. A missing entry does NOT hide an agent.
>
- > **And it never sets the row's colour.** SUCCESS/FAILED is decided by
- > `agent/agent_verdict.py` (v1.48.2) from the agent's OWN `INI_SECTION`
- > self-report, which **OUTRANKS the process exit code**. If a row is coloured
- > wrong, fix the `status:` the agent emits — a read-only diagnostic reporting a
- > finding (`invalid`, `findings`, `no_matches`, `listed`, …) must be a SUCCESS
- > and must `sys.exit(0)`; only `refused` / `not_found` / `engine_unavailable` /
- > `error` / `failed` are red. **Never** add a special case to `mcp_agent.py`, and
- > **never** re-inline a second copy of `DIAGNOSTIC_COMPLETED_STATUSES` — there is
- > exactly ONE definition, in `agent_verdict.py`, and a drifted copy silently
- > mis-colours rows.
+ > **And it never sets the row's colour.** SUCCESS/FAILED is decided by
+ > `agent/agent_verdict.py` from the agent's OWN `INI_SECTION` self-report, which
+ > **OUTRANKS the process exit code**. Its CLOSED vocabulary has five disjoint
+ > sets: diagnostic-completed and work-completed statuses are green; degraded,
+ > work-not-done, and agent-error statuses are red. A completed read-only
+ > diagnostic must still `sys.exit(0)`. Never write a numeric return code in
+ > `status:`; emit `returncode`, `success`, and a semantic status token. **Never**
+ > add a special case to `mcp_agent.py` or re-inline any vocabulary set — there
+ > is exactly ONE definition in `agent_verdict.py`, guarded by
+ > `agent/test_status_vocabulary.py`.
Three-step procedure (matches `docs/claude/exec-report.md`):
1. Add an entry to `_EXEC_REPORT_TOOLS` in `Tlamatini/agent/mcp_agent.py`:
```python
"${input.tool_name}": ("${input.agent_key}", "${input.agent_display}"),
```
2. Add CSS rules in `agent_page.css`:
- `.exec-report-caption-${input.agent_key}` (gradient mirroring
`.canvas-item.${input.agent_key}-agent` from agentic_control_panel.css)
- `.exec-report-${input.agent_key} .exec-report-cmd { border-left: 3px solid <primary>; }`
- If the caption is dark, append `.exec-report-${input.agent_key} thead th`
to the dark-tinted override selector list.
- 3. Run `python Tlamatini/manage.py test agent.tests.ExecReportCaptureTests`.
+ 3. Run `python Tlamatini/manage.py test agent.tests.ExecReportCaptureTests agent.test_agent_verdict agent.test_exec_report_verdict agent.test_status_vocabulary`.
Report the pass/fail status. If the verdict/colour logic was touched at all,
also run `agent.test_agent_verdict` and `agent.test_exec_report_verdict`.
Return `{ rows_added, tests_pass }`.