de-aigc-skills

Each tab says whether it installs commit 371a40e, the one we checked, or the latest version.

tmp="$(mktemp -d)" &&
  git clone -q --depth 1 --filter=blob:none --no-checkout https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills.git "$tmp" &&
  git -C "$tmp" fetch -q --depth 1 origin 371a40e488c5d9e8e75ca2af5dd9667f8349c199 &&
  git -C "$tmp" sparse-checkout set -- skills/48-de-AIGC-skills &&
  git -C "$tmp" checkout -q 371a40e488c5d9e8e75ca2af5dd9667f8349c199 &&
  mkdir -p .claude/skills &&
  mkdir .claude/skills/de-aigc-skills &&
  cp -R "$tmp"/skills/48-de-AIGC-skills/. .claude/skills/de-aigc-skills/ &&
  rm -rf "$tmp"

tmp="$(mktemp -d)" &&
  git clone -q --depth 1 --filter=blob:none --no-checkout https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills.git "$tmp" &&
  git -C "$tmp" fetch -q --depth 1 origin 371a40e488c5d9e8e75ca2af5dd9667f8349c199 &&
  git -C "$tmp" sparse-checkout set -- skills/48-de-AIGC-skills &&
  git -C "$tmp" checkout -q 371a40e488c5d9e8e75ca2af5dd9667f8349c199 &&
  out="$(mktemp -d)" &&
  cp -R "$tmp"/skills/48-de-AIGC-skills "$out"/de-aigc-skills &&
  (cd "$out" && zip -qr - de-aigc-skills) > de-aigc-skills.zip &&
  rm -rf "$tmp" "$out"

tmp="$(mktemp -d)" &&
  git clone -q --depth 1 --filter=blob:none --no-checkout https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills.git "$tmp" &&
  git -C "$tmp" fetch -q --depth 1 origin 371a40e488c5d9e8e75ca2af5dd9667f8349c199 &&
  git -C "$tmp" sparse-checkout set -- skills/48-de-AIGC-skills &&
  git -C "$tmp" checkout -q 371a40e488c5d9e8e75ca2af5dd9667f8349c199 &&
  mkdir -p .agents/skills &&
  mkdir .agents/skills/de-aigc-skills &&
  cp -R "$tmp"/skills/48-de-AIGC-skills/. .agents/skills/de-aigc-skills/ &&
  rm -rf "$tmp"

tmp="$(mktemp -d)" &&
  git clone -q --depth 1 --filter=blob:none --no-checkout https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills.git "$tmp" &&
  git -C "$tmp" fetch -q --depth 1 origin 371a40e488c5d9e8e75ca2af5dd9667f8349c199 &&
  git -C "$tmp" sparse-checkout set -- skills/48-de-AIGC-skills &&
  git -C "$tmp" checkout -q 371a40e488c5d9e8e75ca2af5dd9667f8349c199 &&
  mkdir -p .cursor/skills &&
  mkdir .cursor/skills/de-aigc-skills &&
  cp -R "$tmp"/skills/48-de-AIGC-skills/. .cursor/skills/de-aigc-skills/ &&
  rm -rf "$tmp"

tmp="$(mktemp -d)" &&
  git clone -q --depth 1 --filter=blob:none --no-checkout https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills.git "$tmp" &&
  git -C "$tmp" fetch -q --depth 1 origin 371a40e488c5d9e8e75ca2af5dd9667f8349c199 &&
  git -C "$tmp" sparse-checkout set -- skills/48-de-AIGC-skills &&
  git -C "$tmp" checkout -q 371a40e488c5d9e8e75ca2af5dd9667f8349c199 &&
  mkdir -p .github/skills &&
  mkdir .github/skills/de-aigc-skills &&
  cp -R "$tmp"/skills/48-de-AIGC-skills/. .github/skills/de-aigc-skills/ &&
  rm -rf "$tmp"

gh skill install brycewang-stanford/Auto-Empirical-Research-Skills skills/48-de-AIGC-skills/SKILL.md@371a40e488c5d9e8e75ca2af5dd9667f8349c199 --agent github-copilot --scope project

tmp="$(mktemp -d)" &&
  git clone -q --depth 1 --filter=blob:none --no-checkout https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills.git "$tmp" &&
  git -C "$tmp" fetch -q --depth 1 origin 371a40e488c5d9e8e75ca2af5dd9667f8349c199 &&
  git -C "$tmp" sparse-checkout set -- skills/48-de-AIGC-skills &&
  git -C "$tmp" checkout -q 371a40e488c5d9e8e75ca2af5dd9667f8349c199 &&
  mkdir -p .gemini/skills &&
  mkdir .gemini/skills/de-aigc-skills &&
  cp -R "$tmp"/skills/48-de-AIGC-skills/. .gemini/skills/de-aigc-skills/ &&
  rm -rf "$tmp"

gemini skills install https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills.git --path skills/48-de-AIGC-skills --scope workspace

tmp="$(mktemp -d)" &&
  git clone -q --depth 1 --filter=blob:none --no-checkout https://github.com/brycewang-stanford/Auto-Empirical-Research-Skills.git "$tmp" &&
  git -C "$tmp" fetch -q --depth 1 origin 371a40e488c5d9e8e75ca2af5dd9667f8349c199 &&
  git -C "$tmp" sparse-checkout set -- skills/48-de-AIGC-skills &&
  git -C "$tmp" checkout -q 371a40e488c5d9e8e75ca2af5dd9667f8349c199 &&
  mkdir -p .agents/skills &&
  mkdir .agents/skills/de-aigc-skills &&
  cp -R "$tmp"/skills/48-de-AIGC-skills/. .agents/skills/de-aigc-skills/ &&
  rm -rf "$tmp"

npx skills@1.7.4 add brycewang-stanford/Auto-Empirical-Research-Skills --skill de-aigc-skills

From the files and frontmatter at commit 371a40e. This is how much a skill can do, not whether it is risky.

What it can touch: Shell, network or credentials

• Runs shell commands

  Its instructions or files give the agent commands to run on your computer.

• Reaches the network

  It sends or fetches data over the internet, for example with curl, an HTTP library or a package install.

• Names web addresses

  It mentions outside web addresses the agent may visit. We list them but don't follow them.

• Bundles scripts

  The folder holds code files the agent may run.

• Pre-approves tools

  Its allowed-tools field lists tools the agent may use without asking you first, in agents that honour it.

  SKILL.md:27

• Changes agent settings

  It writes to agent memory or settings files such as CLAUDE.md, AGENTS.md or .claude/settings.json.

  README.md:115 · README.md:118

Pattern check: 7 items to review · commit 371a40e · 27 Oct 2026

• R34 · HighBroad tools pre-approved in allowed-tools

  SKILL.md:27 (opens in a new tab)

  - Bash(python3 scripts/provenance_scrub.py:*)

• R19 · HighInvisible or direction-changing characters

  references/watermarks.md:113 (opens in a new tab)

  pattern scans and detector n-grams see `re⟨U+200B⟩form` as two tokens. Clean first,

• R19 · HighInvisible or direction-changing characters

  scripts/provenance_scrub.py:1087 (opens in a new tab)

  ("zwsp removed", "en", "The re⟨U+200B⟩form reduced entry.", "The reform reduced entry."),

• R19 · HighInvisible or direction-changing characters

  scripts/provenance_scrub.py:1089 (opens in a new tab)

  ("word joiner removed", "en", "Table⟨U+2060⟩3", "Table3"),

• R19 · HighInvisible or direction-changing characters

  scripts/provenance_scrub.py:1097 (opens in a new tab)

  ("bidi stripped in latin", "en", "abc⟨U+202E⟩def⟨U+202C⟩", "abcdef"),

• R19 · HighInvisible or direction-changing characters

  scripts/provenance_scrub.py:1100 (opens in a new tab)

  ("zwnj stripped in latin", "en", "ab⟨U+200C⟩cd", "abcd"),

• R19 · HighInvisible or direction-changing characters

  scripts/provenance_scrub.py:1103 (opens in a new tab)

  ("zwj stripped in prose", "en", "fir⟨U+200D⟩m", "firm"),

We checked every text file in this skill, at commit 371a40e on 27 Oct 2026, against 45 published patterns such as download-and-run commands, hidden text and credential access. We don't run the skill, we don't read the web pages it points to, and we can't judge everything its instructions might lead an agent to do. In published tests, pattern checks like this one missed most malicious skills. Read the skill before you install it — especially anything it asks you to download or run.

How the check worksReport an incorrect finding

Each agent reads SKILL.md a little differently. We check the file against each one's published rules; we don't run the skill in any agent.

• Agent Skills specFails

  The open format from agentskills.io, held to what the strictest agents enforce, so one folder loads everywhere.

  • Error

    Line 3: the description value contains a colon followed by a space, which YAML reads as the start of another field. Strict YAML parsers, including the spec's validator, reject the file. F4

    Fix: Put the value in quotes, like this: description: '中英双语学术降 AIGC / bilingual academic de-AIGC skill. Removes AI-generated writing signatu…

  • Error

    The name "de-aigc-skills" doesn't match its folder "48-de-AIGC-skills". The spec requires them to match. N4

    Fix: Set name: 48-de-aigc-skills, or rename the folder to de-aigc-skills.

  • Warning

    Fields outside the spec: triggers. Uploads to the Claude apps and API, and the spec's validator, reject the file; most agents ignore these fields. O1

    Fix: Delete them for a portable skill, or keep them if you only target the agents that read them.

  • Warning

    allowed-tools is a list; the spec defines a space-separated string. gh skill publish rejects a list, though Claude Code accepts one. O5

    Fix: Write it on one line: allowed-tools: Bash(python3 scripts/provenance_scrub.py:*)

  • Warning

    Paths that may not work once the skill is installed: line 299: ../44-matsuikentaro1-humanizer_academic/ (outside the skill folder), line 301: ../45-stephenturner-skill-deslop/ (outside the skill folder), line 302: ../46-hardikpandya-stop-slop/ (outside the skill folder), line 304: ../47-conorbronsdon-avoid-ai-writing/ (outside the skill folder), line 306: ../49-voidborne-d-humanize-chinese/ (outside the skill folder) and 1 more. B6

    Fix: Use forward-slash paths relative to SKILL.md, and copy shared files into the skill.

  • Note

    There is no license field, so readers can't tell whether they may reuse the skill. O2

    Fix: Add license: with an SPDX name such as MIT or Apache-2.0, or the name of a bundled licence file.

  • Note

    Reference files over 100 lines with no table of contents at the top: references/examples-en.md (199 lines), references/examples-zh.md (217 lines), references/patterns-en.md (293 lines), references/patterns-zh.md (231 lines), references/scoring.md (207 lines) and 2 more. B5

    Fix: Add a short list of the sections at the top of each.

  • Note

    Line 72 has dated wording ("Since 2026"), which goes stale. B8

    Fix: Describe the current way, and move old patterns into their own clearly marked section.

• Claude CodePasses with warnings

  Claude Code reads its own extra fields, ignores unknown ones, and still loads a skill whose frontmatter is broken, without its fields.

  • Warning

    Paths that may not work once the skill is installed: line 299: ../44-matsuikentaro1-humanizer_academic/ (outside the skill folder), line 301: ../45-stephenturner-skill-deslop/ (outside the skill folder), line 302: ../46-hardikpandya-stop-slop/ (outside the skill folder), line 304: ../47-conorbronsdon-avoid-ai-writing/ (outside the skill folder), line 306: ../49-voidborne-d-humanize-chinese/ (outside the skill folder) and 1 more. B6

    Fix: Use forward-slash paths relative to SKILL.md, and copy shared files into the skill.

  • Note

    Line 3: the description value contains a colon followed by a space, which YAML reads as the start of another field. Claude Code quotes the value itself and loads the skill, but other agents reject it. F4

    Fix: Put the value in quotes, like this: description: '中英双语学术降 AIGC / bilingual academic de-AIGC skill. Removes AI-generated writing signatu…

  • Note

    The name "de-aigc-skills" doesn't match its folder "48-de-AIGC-skills". Claude Code shows the skill as /de-aigc-skills, and the folder name also invokes it. N4

    Fix: Set name: 48-de-aigc-skills, or rename the folder to de-aigc-skills.

  • Note

    Claude Code ignores fields it doesn't know: triggers. O1

    Fix: Check the spelling against Claude Code's field list, or delete them.

  • Note

    Reference files over 100 lines with no table of contents at the top: references/examples-en.md (199 lines), references/examples-zh.md (217 lines), references/patterns-en.md (293 lines), references/patterns-zh.md (231 lines), references/scoring.md (207 lines) and 2 more. B5

    Fix: Add a short list of the sections at the top of each.

  • Note

    Line 72 has dated wording ("Since 2026"), which goes stale. B8

    Fix: Describe the current way, and move old patterns into their own clearly marked section.

• Claude apps & API uploadFails

  Uploading to claude.ai or the Claude API, or packaging with Anthropic's script, accepts only the six spec fields and stops on anything else.

  • Error

    Line 3: the description value contains a colon followed by a space, which YAML reads as the start of another field. Anthropic's validator rejects it, so packaging and upload fail. F4

    Fix: Put the value in quotes, like this: description: '中英双语学术降 AIGC / bilingual academic de-AIGC skill. Removes AI-generated writing signatu…

  • Error

    The name "de-aigc-skills" doesn't match its folder "48-de-AIGC-skills". claude.ai refuses an upload whose folder name doesn't match the skill name. N4

    Fix: Set name: 48-de-aigc-skills, or rename the folder to de-aigc-skills.

  • Error

    Upload and packaging fail on fields outside the spec: triggers. O1

    Fix: Delete them, and keep your own data under metadata as text values.

  • Warning

    Paths that may not work once the skill is installed: line 299: ../44-matsuikentaro1-humanizer_academic/ (outside the skill folder), line 301: ../45-stephenturner-skill-deslop/ (outside the skill folder), line 302: ../46-hardikpandya-stop-slop/ (outside the skill folder), line 304: ../47-conorbronsdon-avoid-ai-writing/ (outside the skill folder), line 306: ../49-voidborne-d-humanize-chinese/ (outside the skill folder) and 1 more. B6

    Fix: Use forward-slash paths relative to SKILL.md, and copy shared files into the skill.

  • Note

    The description is 942 characters. The claude.ai Help Center says 200 at most, though Anthropic's API docs and validator allow 1,024. D4

    Fix: If the claude.ai upload refuses it, trim the description to 200 characters.

  • Note

    allowed-tools is a list; the spec defines a space-separated string. O5

    Fix: Write it on one line: allowed-tools: Bash(python3 scripts/provenance_scrub.py:*)

  • Note

    Reference files over 100 lines with no table of contents at the top: references/examples-en.md (199 lines), references/examples-zh.md (217 lines), references/patterns-en.md (293 lines), references/patterns-zh.md (231 lines), references/scoring.md (207 lines) and 2 more. B5

    Fix: Add a short list of the sections at the top of each.

  • Note

    Line 72 has dated wording ("Since 2026"), which goes stale. B8

    Fix: Describe the current way, and move old patterns into their own clearly marked section.

• OpenAI (Codex & API)Fails

  Codex finds SKILL.md by its exact name and needs a description; OpenAI's Skills API checks the frontmatter against the spec and caps uploads.

  • Error

    Line 3: the description value contains a colon followed by a space, which YAML reads as the start of another field. Codex quotes the value and loads the skill, but OpenAI's upload follows the spec, which needs valid YAML. F4

    Fix: Put the value in quotes, like this: description: '中英双语学术降 AIGC / bilingual academic de-AIGC skill. Removes AI-generated writing signatu…

  • Warning

    The name "de-aigc-skills" doesn't match its folder "48-de-AIGC-skills". OpenAI's own upload example uses a differently named folder, but the spec it follows says they must match. N4

    Fix: Set name: 48-de-aigc-skills, or rename the folder to de-aigc-skills.

  • Warning

    Fields outside the spec: triggers. Codex ignores them; OpenAI doesn't say whether its upload accepts them. O1

    Fix: Delete them, or move your own data under metadata as text values.

  • Warning

    Paths that may not work once the skill is installed: line 299: ../44-matsuikentaro1-humanizer_academic/ (outside the skill folder), line 301: ../45-stephenturner-skill-deslop/ (outside the skill folder), line 302: ../46-hardikpandya-stop-slop/ (outside the skill folder), line 304: ../47-conorbronsdon-avoid-ai-writing/ (outside the skill folder), line 306: ../49-voidborne-d-humanize-chinese/ (outside the skill folder) and 1 more. B6

    Fix: Use forward-slash paths relative to SKILL.md, and copy shared files into the skill.

  • Note

    allowed-tools is a list; the spec defines a space-separated string. O5

    Fix: Write it on one line: allowed-tools: Bash(python3 scripts/provenance_scrub.py:*)

• VS Code & CopilotPasses with warnings

  VS Code reads skills one folder below .github/skills, .claude/skills or .agents/skills and falls back to the folder name when the name is missing or differs; GitHub's Copilot docs require the exact name SKILL.md.

  • Warning

    Line 3: the description value contains a colon followed by a space, which YAML reads as the start of another field. VS Code's own YAML reader still loads the skill, but may read some values differently. F4

    Fix: Put the value in quotes, like this: description: '中英双语学术降 AIGC / bilingual academic de-AIGC skill. Removes AI-generated writing signatu…

  • Warning

    The name "de-aigc-skills" doesn't match its folder "48-de-AIGC-skills". VS Code uses the folder name, 48-de-AIGC-skills, instead. N4

    Fix: Set name: 48-de-aigc-skills, or rename the folder to de-aigc-skills.

  • Warning

    Paths that may not work once the skill is installed: line 299: ../44-matsuikentaro1-humanizer_academic/ (outside the skill folder), line 301: ../45-stephenturner-skill-deslop/ (outside the skill folder), line 302: ../46-hardikpandya-stop-slop/ (outside the skill folder), line 304: ../47-conorbronsdon-avoid-ai-writing/ (outside the skill folder), line 306: ../49-voidborne-d-humanize-chinese/ (outside the skill folder) and 1 more. B6

    Fix: Use forward-slash paths relative to SKILL.md, and copy shared files into the skill.

  • Note

    VS Code ignores fields it doesn't support: triggers. O1

    Fix: Delete them unless another agent you target reads them.

  • Note

    allowed-tools is a list; the spec defines a space-separated string. O5

    Fix: Write it on one line: allowed-tools: Bash(python3 scripts/provenance_scrub.py:*)

10 files