Anti Defensive Writing En Agent Skill

The Press-Release Principle (Paper as Press Conference)

One-line principle

A paper is a press conference, not a project summary, a lab log, or a self-audit.

Your job is not to present every aspect of the work evenly, but to identify its

most publishable value and build the most favorable, complete, and persuasive

narrative around it.

1. Narrative rules

• Organize the paper around strengths only. Look for what is genuinely ahead,

unique, or irreplaceable: a new capability, a new problem, a new mechanism, a new

perspective, broader applicability, lower cost, higher efficiency, better

scalability, or a more meaningful trade-off. Content that forms no advantage

does not enter the main storyline.

• Do not write a work report. No "we first did X, then tried Y, and Z failed"

chronology. Keep only the final standing logic: why the problem matters → why

existing methods fall short → what this paper provides → how the evidence

supports it.

• Never set a contest you cannot win. If a metric is not your strength, do not

build the narrative on it. Re-choose the task definition, evaluation dimension,

application scenario, constraint, or comparison frame that best reflects your

value. Name the game this paper actually wins.

• State the advantage explicitly. Do not expect reviewers to discover the

contribution from a table. Explain: under which condition the method performs

best, why the advantage emerges, what practical problem it solves, and why it

deserves attention.

• Limit the comparison scope. Do not chase "wins on every dataset and metric."

Make only claims your evidence firmly supports. Persuasion comes from tight

claim-evidence alignment, not from the number of comparisons.

• Allow full story restructuring. When existing results cannot support the

original narrative, do not defend it. Redefine the problem, re-order the

contributions, re-pick the headline result, redesign title, abstract,

introduction, and experiment structure. The story serves the strongest

evidence, not the original plan.

2. Language rules (against self-undermining)

• Banned self-weakening phrases: "unfortunately", "merely", "only achieves",

"still lags far behind", "limited improvement", "severely insufficient".

• Never say you lost. For unfavorable results, avoid "our method is weaker",

"performance drops", "fails to surpass". Judge in order:

① must this result be discussed at all; ② can it be explained by different

goals, constraints, or applicable scenarios; ③ does it reflect a reasonable

trade-off; ④ can the claim be narrowed to avoid a head-on collision;

⑤ should the storyline change so this result no longer carries the core

contribution.

• If a result forms no advantage, do not distill it into a conclusion — and never

help the reader escalate a local observation into a verdict on the whole method.

3. Experiment rules

Experiments are tools of argument, not a warehouse of results. Every experiment

must carry one explicit duty:

• prove the core method works;
• prove the advantage comes from the key mechanism;
• prove the method matters in the target scenario;
• rule out the most plausible alternative explanations.

An experiment that does not strengthen the main line, distracts, or invites

irrelevant disputes: delete, downplay, move, or redesign it.

4. Structure rules

Abstract and introduction = the opening of the press conference. Establish

four things fast: ① an important, unsolved problem; ② the key gap in existing

methods; ③ your distinctive approach; ④ your strongest result and its meaning.

Do not open with implementation details, process, or a wall of background. Do not

discuss limitations before the contribution is established.

The conclusion only reinforces the takeaway: what was solved, what was

proposed, what was proven, why it matters. Never introduce a fresh self-negation

or expanded limitations in the final paragraph.

5. Default decision rules (for any unfavorable material, in priority order)

• Delete content irrelevant to the core claim;
• Narrow the claim to avoid a pointless head-on comparison;
• Switch to an evaluation dimension that better reflects the value;
• Explain the result as a difference in goals or a reasonable trade-off;
• Reorganize experiments so the advantage becomes the visual and narrative center;
• Redefine the paper's story;
• Only when unavoidable and truly affecting the core conclusion, state it plainly.

6. Pre-submission checklist (never hand the reviewer a knife)

• Does this sentence quietly expand the responsibility the paper must carry?
• Does it raise a question nobody asked for?
• Does it describe a local phenomenon as a universal flaw?
• Does it use a negative judgment broader than the evidence?
• Is any passage unfolding as "what we did" process narration?
• Is a non-winning metric set as the main battlefield?
• In the first five sentences of the abstract, are both the problem and the

contribution present?

• Does the conclusion smuggle in a sudden self-negation?

Do not manufacture review questions. Do not enlarge the attack surface.

Do not complete the opposing side's argument for them.

Final goal

Every section, paragraph, table, and sentence should do one thing together:

convince the reader that this work solves a problem worth solving, proposes a

method worth noticing, and already has sufficiently clear evidence of its value.

Find the advantage that truly holds, organize all material around it, and make

that advantage unmistakably clear.