hermes-agent/hermes_cli
teknium1 a8bf414f4a feat: browser console/errors tool, annotated screenshots, auto-recording, and dogfood QA skill
New browser capabilities and a built-in skill for agent-driven web QA.

## New tool: browser_console

Returns console messages (log/warn/error/info) AND uncaught JavaScript
exceptions in a single call. Uses agent-browser's 'console' and 'errors'
commands through the existing session plumbing. Supports --clear to reset
buffers. Verified working in both local and Browserbase cloud modes.

## Enhanced tool: browser_vision(annotate=True)

New boolean parameter on browser_vision. When true, agent-browser overlays
numbered [N] labels on interactive elements — each [N] maps to ref @eN.
Annotation data (element name, role, bounding box) returned alongside the
vision analysis. Useful for QA reports and spatial reasoning.

## Config: browser.record_sessions

Auto-record browser sessions as WebM video files when enabled:
- Starts recording on first browser_navigate
- Stops and saves on browser_close
- Saves to ~/.hermes/browser_recordings/
- Works in both local and cloud modes (verified)
- Disabled by default

## Built-in skill: dogfood

Systematic exploratory QA testing for web applications. Teaches the agent
a 5-phase workflow:
1. Plan — accept URL, create output dirs, set scope
2. Explore — systematic crawl with annotated screenshots
3. Collect Evidence — screenshots, console errors, JS exceptions
4. Categorize — severity (Critical/High/Medium/Low) and category
   (Functional/Visual/Accessibility/Console/UX/Content)
5. Report — structured markdown with per-issue evidence

Includes:
- skills/dogfood/SKILL.md — full workflow instructions
- skills/dogfood/references/issue-taxonomy.md — severity/category defs
- skills/dogfood/templates/dogfood-report-template.md — report template

## Tests

21 new tests covering:
- browser_console message/error parsing, clear flag, empty/failed states
- browser_console schema registration
- browser_vision annotate schema and flag passing
- record_sessions config defaults and recording lifecycle
- Dogfood skill file existence and content validation

Addresses #315.
2026-03-08 21:28:12 -07:00
..
__init__.py Hermes Agent UX Improvements 2026-02-22 02:16:11 -08:00
auth.py fix: add Kimi Code API support (api.kimi.com/coding/v1) 2026-03-07 21:00:12 -05:00
banner.py feat: show update-available notice in CLI banner 2026-03-07 07:35:36 -08:00
callbacks.py refactor: reorganize agent and CLI structure for improved clarity 2026-02-21 23:17:18 -08:00
clipboard.py fix: clipboard BMP conversion file loss and broken test 2026-03-08 17:22:27 -07:00
codex_models.py fix: trust user-selected models with OpenAI Codex provider 2026-03-08 18:29:09 -07:00
colors.py Cleanup time! 2026-02-20 23:23:32 -08:00
commands.py feat: session naming with unique titles, auto-lineage, rich listing, resume by name 2026-03-08 15:20:29 -07:00
config.py feat: browser console/errors tool, annotated screenshots, auto-recording, and dogfood QA skill 2026-03-08 21:28:12 -07:00
cron.py refactor: streamline cron job handling and update CLI commands 2026-02-21 16:21:19 -08:00
doctor.py fix: add Kimi Code API support (api.kimi.com/coding/v1) 2026-03-07 21:00:12 -05:00
gateway.py feat: add Signal messenger gateway platform (#405) 2026-03-08 20:20:35 -07:00
main.py feat: interactive session browser with search filtering (#718) 2026-03-08 17:42:50 -07:00
models.py feat: /provider command + fix gateway bugs + harden parse_model_input 2026-03-08 06:09:36 -07:00
pairing.py Cleanup time! 2026-02-20 23:23:32 -08:00
runtime_provider.py feat: add z.ai/GLM, Kimi/Moonshot, MiniMax as first-class providers 2026-03-06 18:55:18 -08:00
setup.py fix(setup): correct import of get_codex_model_ids in setup wizard 2026-03-08 13:07:19 -07:00
skills_hub.py fix: Initialize Skills Hub on list 2026-03-09 01:43:59 +08:00
status.py fix: update setup/config UI for local browser mode 2026-03-07 01:23:27 -08:00
tools_config.py Fix empty tool selection persistence 2026-03-08 06:11:18 -07:00
uninstall.py Cleanup time! 2026-02-20 23:23:32 -08:00