Agent skill

Press Conference Revision Evidence Bound

by lensback940701 in lensback940701/Evidence-Bound-Press-Conference-Revision-Skill

Diagnose and revise defensive academic writing while preserving claim ceilings, evidence status, scope conditions, rival explanations, and conceptual hierarchy.

MITAuto-check passedDocuments & Office

Install Press Conference Revision Evidence Bound

skills CLI
$ npx skills add lensback940701/Evidence-Bound-Press-Conference-Revision-Skill --skill press-conference-revision-evidence-bound -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install lensback940701/Evidence-Bound-Press-Conference-Revision-Skill press-conference-revision-evidence-bound --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
press-conference-revision-evidence-bound
GitHub stars
259
Token cost
~2.2k tokens
SKILL.md length
1,077 words
Files
12 (incl. scripts, references, assets)
Skills in repo
1
Repo updated
First seen
Licence
MIT

At a glance

Diagnose and revise defensive academic writing while preserving claim ceilings, evidence status, scope conditions, rival explanations, and conceptual hierarchy.

  • Works in 4 steps: create a contribution-and-scope contract → batch diagnostic (read-only) → concentrated revision (only with… → …
  • A whole-manuscript batch audit
  • SKILL.md covers Non-negotiable distinction, Required inputs and authority…, Phase 0 — create a… and Phase 1 — batch diagnostic…, plus 4 more sections
  • Runs Python scripts from its folder

What it does

Press Conference Revision Evidence Bound is an agent skill from lensback940701/Evidence-Bound-Press-Conference-Revision-Skill. Diagnose and revise defensive academic writing while preserving claim ceilings, evidence status, scope conditions, rival explanations, and conceptual hierarchy. Use for a whole-manuscript batch audit or a tightly authorized, tracked-change revision when an academic paper hides its contribution beneath self-defence, repeated caveats, process narration, or unnecessary comparisons; especially when a qualitative, historical, or mixed-method manuscript must become contribution-forward without becoming promotional or…

Its SKILL.md is about 2.2k tokens, which your agent loads only when the skill is triggered. The skill folder holds 16 other files, including scripts, reference files and assets (for example `README.md`, `README.zh-CN.md` and `agents/openai.yaml`).

It sits in Documents & Office, covering Text to speech and voice, Word documents and Citation management. The repository describes itself as: An evidence-bound academic revision skill for social science research. It reduces defensive prose without weakening scholarly caution, foregrounds core contributions, compresses… The licence is MIT.

When your agent uses it

  • A whole-manuscript batch audit
  • A tightly authorized
  • Tracked-change revision when an academic paper hides its contribution beneath self-defence
  • Repeated caveats

Example prompts

  • “/press-conference-revision-evidence-bound”

Requirements

  • Python 3

Workflow steps

4 steps, taken from the step headings in SKILL.md.

  1. create a contribution-and-scope contract
  2. batch diagnostic (read-only)
  3. concentrated revision (only with authority)
  4. whole-manuscript regression

What it can do on your machine

Read from SKILL.md and the folder at commit 33e6c3a. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships 1 file in scripts/ (Python), which the agent can run.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Press Conference Revision Evidence Bound loads about 2.2k tokens when it runs, and up to ~4k if it reads all its reference files. Until then it costs about 184 tokens; SKILL.md has 1,077 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~184
When it runs · the whole SKILL.md, loaded when a task matches
~2.2k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~4k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from lensback940701/Evidence-Bound-Press-Conference-Revision-Skill at commit 33e6c3a, republished under its MIT licence (© lensback940701). 1,077 words, ~2,243 tokens.

Download SKILL.mdSave it as .claude/skills/press-conference-revision-evidence-bound/SKILL.md (or your agent's skills folder). This skill also uses 11 other files; get the full folder from GitHub.
name
press-conference-revision-evidence-bound
description
Diagnose and revise defensive academic writing while preserving claim ceilings, evidence status, scope conditions, rival explanations, and conceptual hierarchy. Use for a whole-manuscript batch audit or a tightly authorized, tracked-change revision when an academic paper hides its contribution beneath self-defence, repeated caveats, process narration, or unnecessary comparisons; especially when a qualitative, historical, or mixed-method manuscript must become contribution-forward without becoming promotional or overclaimed. Do not use for first drafting, new research, citation repair, evidence admission, grammar-only polishing, or making claims sound stronger than the evidence permits.

Evidence-Bound Press-Conference Revision

Treat a paper as a clear academic presentation of its best supported contribution, not as a work log or a self-audit. Reframe defensiveness; never remove the information that lets readers judge what the evidence supports.

Non-negotiable distinction

Do not treat a cue word as a verdict. The target is rhetorical defensiveness, not scholarly caution.

  • Cut, compress, or reframe self-deprecation, anticipatory answers to objections the manuscript has not invited, redundant hedge stacks, process-log narration, and negative judgments broader than the evidence.
  • Keep or relocate a real scope condition, source-status distinction, rival explanation, contradiction, negative finding, ethical caveat, or method limitation whenever it changes how the claim must be read.
  • When uncertain, retain the wording and register a query. Do not convert possible to certain, report to observation, authorization to delivery, sequence to mechanism, or a bounded case result to a general law.

Read references/triage-and-regression.md before classifying candidates. For the Dalian Urban Geography manuscript, also read references/urban-geography-guardrails.md before any diagnosis or revision.

Required inputs and authority order

Collect the smallest current set of materials needed for the task:

  1. the latest clean manuscript and, for a revision pass, its tracked baseline;
  2. the current manuscript lock and any later author-authorized departure from it;
  3. current author decisions and unresolved queries;
  4. target-journal guidance and any supplied corpus/style calibration;
  5. the source principles and prior audit/return records, as context rather than evidence.

Apply this order: manuscript evidence and active author decisions → lock and claim ceiling → verified journal guidance → corpus-level rhetorical calibration → this skill. Do not use a corpus as a phrase bank or let a style preference override a source status.

Phase 0 — create a contribution-and-scope contract

Before scanning, write a compact contract. Do not edit until it is complete.

fieldrequired content
core contributionOne evidence-bound sentence naming the primary analytical move and payoff.
contribution hierarchyPrimary concept; supporting lens; bounded empirical contribution; portable payoff, if any.
claim ceilingThe strongest permissible formulation already supported by the manuscript and its evidence.
load-bearing cautionsScope, source-status, method, rival, contradiction, and ethics statements that must survive.
edit authoritySections/paragraphs authorized, locked items, tracking requirements, and prohibited changes.
live queriesMeaning-dependent choices that must not be silently resolved.

If the hierarchy, evidence ceiling, or editing authority is unclear, stop after the diagnostic register and request a decision.

Phase 1 — batch diagnostic (read-only)

Read the whole manuscript once for section roles and the contribution contract. Then make a candidate register; do not revise prose in this phase.

  1. Optionally run scripts/scan_defensive_cues.py on the clean DOCX to create a candidate list only. Inspect every result in context; the script makes no rhetorical or evidentiary decision.
  2. Review every prose-bearing paragraph, including title, abstract, introductions, discussion, conclusion, captions, notes, and references only where they affect visible prose.
  3. Classify each candidate using the taxonomy in references/triage-and-regression.md.
  4. Record the candidate in a copy of assets/defensive-writing-audit-template.csv or an equivalent table. Include its evidentiary function, source-status tag, contribution relation, and provisional disposition.
  5. Mark non-defensive but load-bearing language as KEEP; this makes the integrity decision auditable rather than invisible.

Allowed dispositions are KEEP, TIGHTEN, REFRAME, RELOCATE, CUT, and QUERY. A CUT must remove rhetoric only, never a factual proposition, citation function, evidence status, rival, or scope limit.

Phase 2 — concentrated revision (only with authority)

Revise by pattern group and section role, not by a blind word hunt. Use tracked changes when the manuscript is under review.

  1. Start with the abstract, introduction, discussion, and conclusion: make the problem, gap, distinctive analytical move, evidence-bound finding, and payoff legible before qualifications appear.
  2. In theory and methods, replace self-defence with precise positioning, but keep definitions, scope, evidence-status distinctions, and analytical rivals.
  3. In empirical sections, retain contradictions, delay, non-adoption, reversal, and negative findings when they are the data or explain the mechanism. Remove only authorial apology or unnecessary rehearsal of a limitation.
  4. Keep a contribution sentence standing in each major section. A later qualifier may narrow it; it may not erase it or turn the paragraph into a response to a hypothetical reviewer.
  5. For every changed sentence, run the three-part preservation test: (a) same or narrower claim, (b) same source-status distinction, and (c) same citation/quotation role.
  6. Route a change to QUERY if it would alter research questions, terminology, hierarchy, evidence, citation role, table/figure text, paragraph order, or a real methodological limitation.

Never solve a mismatch by adding evidence, strengthening a causal verb, or extending a case-specific statement into a general claim.

Show full SKILL.md (332 more words)Show less

Phase 3 — whole-manuscript regression

Read the revised manuscript independently of the change list. Use the regression protocol in references/triage-and-regression.md.

Verify all of the following:

  • contribution hierarchy is stable from title and abstract through conclusion;
  • a new general-first opening is aligned with the body, or the mismatch is explicitly left as a query;
  • every changed claim stays within the original evidence ceiling;
  • plan, authorization, reported work, observation, outcome, interpretation, and causal inference remain distinct;
  • terminology is consistent without flattening meaningful distinctions;
  • citations, quotations, dates, numbers, captions, footnotes, figure/table cross-references, and paragraph boundaries remain protected;
  • no automatic global replacement has elevated a secondary lens or introduced a prohibited keyword;
  • the clean file has accepted changes only and the tracked file exposes every text change.

Use a global terminology scan only to locate candidates. Never replace a concept merely because it is frequent.

Dalian Urban Geography special gate

When the target is the Dalian heritage-metabolism manuscript, apply the project guardrails before Phase 1 and again before delivery. In particular:

  • keep heritage metabolism primary and operation-centred;
  • keep rescaling supporting, rival-open, and conditional on a documented capacity-to-operation relation;
  • do not promote governance/governing into a master concept or weaken source-specific terms by global substitution;
  • preserve overlapping configurations, contradiction, lag, decoupling, reversal, and recombination as analytic findings rather than defensive residue;
  • treat the recent general-first abstract/introduction reframe as a cross-section consistency question, not permission to rewrite the body without explicit authority.

Deliverables and stop rules

For a diagnostic pass, return:

  • the contribution-and-scope contract;
  • the candidate register with KEEP decisions visible;
  • a section-level concentration map and an exception/query register;
  • a bounded revision recommendation, including which changes require author authority.

For an authorized revision pass, additionally return:

  • tracked and clean manuscript copies;
  • a change summary organized by pattern group;
  • a claim/evidence regression report;
  • a list of deliberate non-edits and unresolved author decisions.

Stop after the requested diagnostic or authorized revision pass. Do not submit, conduct new research, change citations, silently resolve a meaning-dependent query, or turn the paper into promotional prose.

© lensback940701, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 11 other files (scripts, references, assets) in the repository root of lensback940701/Evidence-Bound-Press-Conference-Revision-Skill.

  • SKILL.md
  • LICENSE
  • README.md
  • README.zh-CN.md
  • agents/openai.yaml
  • assets/defensive-writing-audit-template.csv
  • assets/readme/audit-loop.png
  • assets/readme/hero.png
  • assets/readme/principle.png
  • references/triage-and-regression.md
  • references/urban-geography-guardrails.md
  • scripts/scan_defensive_cues.py

Open the folder on GitHubat commit 33e6c3a

Compare with similar skills

Press Conference Revision Evidence Bound next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Press Conference Revision Evidence Bound compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Press Conference Revision Evidence Bound this skilllensback940701/Evidence-Bound-Press-Conference-Revision-Skill259—~2.2kAutomated safety check: PassMIT
Review Papermaxwell2732/my-submission-formatting-agent111—~1.2kAutomated safety check: PassNone
Bibliographyaipoch/medical-research-skills1.9k—~1.6kAutomated safety check: PassMIT
Journal Copyeditor DOCXmikemikeqqq/copyeditor-skill292—~3.9kAutomated safety check: PassMIT
Latex To Word Workflowhajimi-kun/latex-to-word-workflow128—~2.3kAutomated safety check: PassMIT
Scholar Auto Researchjoshzyj/open-scholar-skill168—~21kAutomated safety check: PassCustom licence

Similar skills

  • Review Paper

    maxwell2732/my-submission-formatting-agent

    Comprehensive manuscript review covering argument structure, methodology, citation completeness, writing quality, and potential referee objections.

    111 GitHub stars~1.2k tokensUpdated 6 mo ago
    Documents & OfficeAuto-check passed
  • Bibliography

    aipoch/medical-research-skills

    Classifies and organizes literature by theme, method, and conclusion; use when you need to batch-read a folder of PDF/MD/DOCX/TXT files and output a structured CSV for literature reviews and…

    1.9k GitHub stars~1.6k tokensUpdated 24 days ago
    Documents & OfficeAuto-check passed
  • Journal Copyeditor DOCX

    mikemikeqqq/copyeditor-skill

    Comprehensive academic journal copyediting and manuscript review for Microsoft Word (.docx) files across disciplines and target journals.

    292 GitHub stars~3.9k tokensUpdated 2 mo ago
    Documents & OfficeAuto-check passed
  • Latex To Word Workflow

    hajimi-kun/latex-to-word-workflow

    Load when the user asks to convert an academic LaTeX manuscript, thesis, or report into an editable Word DOCX, adapt it to a Word template, preserve citations or cross-references, or diagnose a…

    128 GitHub stars~2.3k tokensUpdated 2 mo ago
    Documents & OfficeAuto-check passed
  • Scholar Auto Research

    joshzyj/open-scholar-skill

    Stable, deterministic social-science research-paper pipeline from idea or data to verified manuscript, citations, replication package, and final md/docx/tex/pdf outputs.

    168 GitHub stars~21k tokensUpdated 22 days ago
    Documents & OfficeAuto-check passed
  • Officecli Academic Paper

    FerroxLabs/wayland

    A skill your agent uses to build academic-style .docx output: journal / conference / thesis chapters carrying formal citation style (APA, Chicago, IEEE, MLA), numbered equations, figure & table…

    608 GitHub starsUsed in 4 repos~11k tokens
    Documents & OfficeAuto-check passed

Questions about Press Conference Revision Evidence Bound

What does Press Conference Revision Evidence Bound do?

Diagnose and revise defensive academic writing while preserving claim ceilings, evidence status, scope conditions, rival explanations, and conceptual hierarchy. Press Conference Revision Evidence Bound is an agent skill from lensback940701/Evidence-Bound-Press-Conference-Revision-Skill. Diagnose and revise defensive academic writing while preserving claim ceilings, evidence status, scope conditions, rival explanations, and conceptual hierarchy.

When should I use Press Conference Revision Evidence Bound?

Press Conference Revision Evidence Bound fits situations like: A whole-manuscript batch audit; A tightly authorized; tracked-change revision when an academic paper hides its contribution beneath self-defence; repeated caveats.

How do I install Press Conference Revision Evidence Bound in Claude Code?

Run `npx skills add lensback940701/Evidence-Bound-Press-Conference-Revision-Skill --skill press-conference-revision-evidence-bound -a claude-code`. Or copy the skill folder (the lensback940701/Evidence-Bound-Press-Conference-Revision-Skill repository) into .claude/skills/press-conference-revision-evidence-bound in your project. Claude Code loads it when a task matches its description.

How do I install Press Conference Revision Evidence Bound in Codex?

Run `npx skills add lensback940701/Evidence-Bound-Press-Conference-Revision-Skill --skill press-conference-revision-evidence-bound -a codex`. Or copy the skill folder (the lensback940701/Evidence-Bound-Press-Conference-Revision-Skill repository) into .agents/skills/press-conference-revision-evidence-bound in your project. Codex loads it when a task matches its description.

Can I use Press Conference Revision Evidence Bound in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add lensback940701/Evidence-Bound-Press-Conference-Revision-Skill --skill press-conference-revision-evidence-bound -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/press-conference-revision-evidence-bound, .gemini/skills/press-conference-revision-evidence-bound, .github/skills/press-conference-revision-evidence-bound and .opencode/skills/press-conference-revision-evidence-bound in your project.

What does Press Conference Revision Evidence Bound need to run?

Going by SKILL.md and its folder, Press Conference Revision Evidence Bound needs Python for the scripts in its folder. Our summary lists: Python 3.

Does Press Conference Revision Evidence Bound access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Press Conference Revision Evidence Bound safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Press Conference Revision Evidence Bound use?

Press Conference Revision Evidence Bound is published under the MIT licence (from the LICENSE file in the skill folder). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Press Conference Revision Evidence Bound use?

About 2.2k tokens (SKILL.md is roughly 9k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 1.8k tokens, read only when the agent opens those files.

What are the alternatives to Press Conference Revision Evidence Bound?

Skills that share tags, products or a category with Press Conference Revision Evidence Bound: Review Paper (maxwell2732/my-submission-formatting-agent, 111 stars), Bibliography (aipoch/medical-research-skills, 1.9k stars), Journal Copyeditor DOCX (mikemikeqqq/copyeditor-skill, 292 stars) and Latex To Word Workflow (hajimi-kun/latex-to-word-workflow, 128 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Press Conference Revision Evidence Bound?

lensback940701 (a GitHub user) maintains it in lensback940701/Evidence-Bound-Press-Conference-Revision-Skill, which has 259 GitHub stars. The repository was last updated on August 23, 2026.

Source: lensback940701/Evidence-Bound-Press-Conference-Revision-Skill on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.