Skip to content

Yila-AI/awesome-defensive-writing / skills/yila-clear-writing

yila-clear-writing

Verified

Diagnose and revise AI defensive padding in drafts, reports, emails, research prose and public explanations when the user requests less hedging or clearer claims. Preserve evidence strength and meaningful reader guidance.

references/technical-en.md

View on GitHub
# Technical guidance

**English** | [简体中文](technical-zh.md)

Choose a task, copy a prompt and add your material. Follow the case link for supplied facts, a revision and the reasoning behind it.

## DW-043 · Beginner tutorial

Make the supplied material clear, direct and informative. Organize it around
the judgment or action the reader needs. Remove repetitive retreat and empty
preambles. Ground facts, numbers, scope, responsibility and commitments in
the material; keep meaningful conditions next to the claims they qualify.
List missing facts separately.

Task: Give the shortest complete path supported by the material. Keep
prerequisites, expected observations and recovery steps; concision must not
remove required actions.

Output: A tutorial with prerequisites, steps, completion checks and recovery
actions.

Input material:
Task: [fill in]
Environment: [fill in]
Steps: [fill in]
Failure cases: [fill in]

Related case: EX-15 in examples-en.md

## DW-044 · Installation instructions

Make the supplied material clear, direct and informative. Organize it around
the judgment or action the reader needs. Remove repetitive retreat and empty
preambles. Ground facts, numbers, scope, responsibility and commitments in
the material; keep meaningful conditions next to the claims they qualify.
List missing facts separately.

Task: Preserve verified commands, versions, scope and order. Explain required
actions without generic warning piles; do not simplify commands by changing
their effects, and label untested environments.

Output: Installation instructions with environment and scope, supplied commands and
completion checks.

Input material:
Verified steps: [fill in]
Versions: [fill in]
Environment: [fill in]
Scope: [fill in]

Related case: EX-15 in examples-en.md

## DW-045 · Troubleshooting

Make the supplied material clear, direct and informative. Organize it around
the judgment or action the reader needs. Remove repetitive retreat and empty
preambles. Ground facts, numbers, scope, responsibility and commitments in
the material; keep meaningful conditions next to the claims they qualify.
List missing facts separately.

Task: Organize symptoms, observations, diagnosis and next action. Separate causes
from hypotheses; avoid blaming the user or defaulting to reinstalling, and
retain data-protection conditions.

Output: A troubleshooting table of symptoms, observations, current hypotheses and
next checks.

Input material:
Symptoms: [fill in]
Logs: [fill in]
Environment: [fill in]
Attempts: [fill in]

Related case: EX-15 in examples-en.md

## DW-046 · SOP prerequisites

Make the supplied material clear, direct and informative. Organize it around
the judgment or action the reader needs. Remove repetitive retreat and empty
preambles. Ground facts, numbers, scope, responsibility and commitments in
the material; keep meaningful conditions next to the claims they qualify.
List missing facts separately.

Task: Preserve order, units, parameters, prohibitions and applicability. Remove
only nonoperational repetition; do not replace a prohibition with a
nonequivalent positive suggestion.

Output: A numbered SOP retaining parameters, prerequisites, sequence and
prohibitions.

Input material:
Original SOP: [fill in]
Scope: [fill in]
Confirmed parameters: [fill in]

Related case: EX-16 in examples-en.md

## DW-047 · Code-review feedback

Make the supplied material clear, direct and informative. Organize it around
the judgment or action the reader needs. Remove repetitive retreat and empty
preambles. Ground facts, numbers, scope, responsibility and commitments in
the material; keep meaningful conditions next to the claims they qualify.
List missing facts separately.

Task: State the trigger, observable consequence and minimal fix, separating
defects from preferences. Do not invent execution results or obscure a real
issue with vague politeness.

Output: Review feedback with trigger, consequence, suggested fix and actual
verification status.

Input material:
Diff: [fill in]
Relevant context: [fill in]
Actual test results: [fill in]

Related case: EX-21 in examples-en.md

## DW-048 · Technical choice

Make the supplied material clear, direct and informative. Organize it around
the judgment or action the reader needs. Remove repetitive retreat and empty
preambles. Ground facts, numbers, scope, responsibility and commitments in
the material; keep meaningful conditions next to the claims they qualify.
List missing facts separately.

Task: Compare options against constraints and comparable evidence. Recommend when
justified and state switching conditions; otherwise identify a useful
verification step.

Output: An option comparison, recommended choice and conditions that would change
it.

Input material:
Constraints: [fill in]
Options: [fill in]
Measurements or docs: [fill in]
Maintenance: [fill in]

Related case: EX-22 in examples-en.md
add-product-e2e-evalpaperclipai/paperclipAdd or extend a Paperclip full-stack runner E2E workflow, fixture, matcher, or report evidence path for local or Daytona execution.Stars: 115kUpdated Claude CodeCodexVerifiedInstall
add-runner-evalpaperclipai/paperclipAdd or extend a Paperclip Runner protocol evaluation definition, roster, assertion, or report fixture with provenance and narrow validation.Stars: 115kUpdated Claude CodeCodexVerifiedInstall
diagnose-why-work-stoppedpaperclipai/paperclipCompatibility entry point for failed, stalled, looping, or over-recovered Paperclip work. Use fix-failed-paperclip-run for the full diagnosis, regression, repair, real verification, and green-PR workflow; honor an explicitly requested diagnosis-only scope.Stars: 115kUpdated Claude CodeCodexVerifiedInstall
fix-failed-paperclip-runpaperclipai/paperclipDiagnose and fix failed, stuck, slow, or incorrectly blocked Paperclip tasks and runs. Use with a task/run URL or failure report to trace the root cause, prove a regression red then green, check second-order effects, verify the real workflow, and deliver a PR with passing checks.Stars: 115kUpdated Claude CodeCodexVerifiedInstall