Skip to content

kklimuk/docx-cli / .claude/skills/weak-agent-test

weak-agent-test

Verified

Run the weak-agent adversarial test harness against docx-cli. Spawns weak exercise agents (Haiku by default, Sonnet to probe, or a local agent harness's pre-produced runs) to perform real document tasks over six scenarios — five editing (MNDA form-fill + font fidelity, invoice table-edit/restructure + logo replace, résumé styling, contract redlining + commenting, contract finalize via accept/reject + comment reply/resolve) and one authoring (T. S. Eliot poetry journal: multi-column, verse, footnotes, links, figure) — renders every result with Word, has opus judge them against ground-truth rubrics, measures each exercise's tool economy, token cost, wall-clock, and correctness (from transcripts for Claude, the exercise.json ledger for the local harness), and synthesizes a prioritized ergonomics report. Use when the user says 'adversarial review', 'test docx-cli with weak agents', 'run the haiku harness', 'weak agent test', or wants to re-run yesterday's adversarial process.

scenarios/contract-finalize/task.md

View on GitHub
# Finalize the returned redline on our freight contract

You represent **Nimbus Freight Systems, Inc.** (the "Company"). Opposing counsel sent
back `contract-redlined.docx` with their tracked changes and margin comments. We need to
respond: accept the changes we agree with, push back on the ones we don't, and put our
position on record for every comment they left. Edit the file directly.

First, look at what they changed and what they flagged. Then work through the specifics
below.

## Tracked changes — accept some, reject others

**Accept** these revisions (the Company agrees):

- **Payment terms** changed from Net 90 → **Net 30**.
- **Non-compete** shortened from five (5) years → **one (1) year**.

**Reject** these revisions (restore the Company's original draft):

- The **liability-cap** change — the Contractor raised the Company's **$100** cap;
  keep the original $100.
- The **deletion of §11 (Personal Guarantee)** — the Company wants the guarantee, so
  reject the deletion (the section stays).

Opposing counsel also **reformatted the indemnification clause** for readability — a
line-spacing change, not a text edit. The Company is fine with the cleaner look —
**accept** it.

Tip: the change list renumbers as you accept or reject, so re-check it as you go, or
handle one change at a time. Each text revision is a deletion paired with an insertion —
act on both halves; the indemnification spacing change is a single entry.

## Comments — reply to all, resolve the addressed ones

Reply to each of the four comments with a one-line Company position (for example,
"Agreed — revised." or "Declined — standard term."). Then resolve the two the Company
has fully addressed:

- **Resolve**: the **auto-renewal** comment and the **arbitration-fees** comment.
- **Leave open**: the **IP-ownership** comment and the **indemnification** comment —
  both are still under negotiation.

## What "done" looks like

The accepted revisions are clean text (no redline left), the rejected ones are back to
the original wording, every comment has a Company reply, and the auto-renewal and
arbitration-fees comments are marked resolved. The IP and indemnification comments
remain open. The indemnification clause shows the Contractor's line-spacing change as
live formatting with no pending-change marker.
finishing-a-development-branchobra/superpowersUse when implementation is complete, all tests pass, and you need to decide how to integrate the workStars: 322kUpdated Claude CodeCodexVerifiedInstall
systematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixesStars: 322kUpdated Claude CodeCodexVerifiedInstall
test-driven-developmentobra/superpowersUse when implementing any feature or bugfix, before writing implementation codeStars: 322kUpdated Claude CodeCodexVerifiedInstall
skill-creatoranthropics/skillsCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.Stars: 197kUpdated Claude CodeCodexVerifiedInstall