Skip to content

Add WorkBuddy desktop setup recipe and 2026-09-12 thinking deviations - #5

Merged
JJasonSun merged 2 commits into
mainfrom
add-workbuddy-setup-and-thinking-deviations
Oct 2, 2026
Merged

JJasonSun merged 2 commits into
mainfrom
add-workbuddy-setup-and-thinking-deviations

Conversation

@JJasonSun

@JJasonSun JJasonSun commented Sep 13, 2026 •

Copy link
Copy Markdown
Owner

Summary

Add a WorkBuddy desktop setup recipe with separate ordinary-text and thinking routes, plus dated ECNU response observations. The configuration follows current model capabilities; desktop implementation details retain their September inspection date.

WorkBuddy models.json
  ecnu-max      → ordinary text
  ecnu-reasoner → thinking, low / high / max
  both         → text-only, conservative context budget

Evidence

  • Before: The Skill had no WorkBuddy recipe or references to these desktop/API observations. After: The recipe documents the array-shaped configuration, startup/editor behavior, request controls, and response checks, with current contracts separated from historical results.
  • All 76 offline tests, repository validation, compilation, skills-ref validation, and whitespace checks passed. The configuration JSON was parsed and checked for route settings and context budgets.
  • Current WorkBuddy behavior and the revised configuration were not live-tested. Dated API observations do not establish current model quality or service availability.

Merge Danger

Door: two-way — revert the documentation changes.

Blast Radius: Docs

No request runner or SDK code changes. Only users following this Skill's WorkBuddy setup and diagnosis guidance are affected.

web-flow and others added 2 commits September 14, 2026 00:35
New file:
- references/workbuddy_setup.md: field-by-field WorkBuddy custom-model
  configuration, including the bare-array format requirement, the startup
  purge cycle, effort field semantics, and runtime verification.

New dated deviations (live-2026-09-12-c, Windows x64, Node.js fetch):
- ecnu-reasoner alias default thinking: bare alias requests activate
  thinking server-side without a client-side thinking parameter.
- ecnu-max reasoning_effort as thinking trigger: reasoning_effort alone
  activates thinking on ecnu-max, contradicting the documented gating.
- Unavailable reasoning effort tiers: minimal and medium return
  intermittent HTTP 500; xhigh is undocumented but stable.
- Rapid-request 401 metadata failure: sub-second sequential requests
  return 401 mimicking auth failure; four-second spacing eliminates it.

Updated routing tables in SKILL.md, README.md, AGENTS.md, workflows.md,
examples.md, and models.md to cross-reference the new recipe and deviations.

All entries follow the existing dated-evidence format and pass
validate_skill.py and the full test suite (69 passed, 1 skipped).
@JJasonSun
JJasonSun merged commit 77faa78 into main Oct 2, 2026
2 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants