Constrained Output Generation

What the source does

• FilesystemNone observedNo reads or writes in the source
• NetworkNone observedNo outbound calls in the source
• ShellNone observedNo commands run in the source
• DependenciesPinnedNothing floats
• SecretsNone foundNo keys in source

Five audit layers

Gates applied: no behavioural pass.

Audit 0.4.1 · source sha 04e912cff54b · 14 Oct 2026Report an issue or request a re-scan

• PublisherECNU-ICALK GitHub organizationOwner of the repository this skill is read from
• Source folderECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt3.5_8_GLM4.7/constrained-output-generationThe exact folder we audited
• LicenceNot declaredAs the repository states it
• Commit pinned for this verdict04e912cff54bA new audit runs when the source changes
• Repository statusNot archivedLast push 157 days ago
• Sandboxed runNot runSkills are read, not executed; layer L4 is skipped

Its description: Restricts all responses to a user-defined list of terms, characters, or numbers until a specific stop phrase is entered by the user.

---
id: "2cae8007-f65c-4fa5-8d1b-c3ee191ca496"
name: "Constrained Output Generation"
description: "Restricts all responses to a user-defined list of terms, characters, or numbers until a specific stop phrase is entered by the user."
version: "0.1.0"
tags:
  - "constraints"
  - "vocabulary"
  - "output control"
  - "game"
  - "restriction"
triggers:
  - "Use only [terms] from now on"
  - "use only numbers from now on"
  - "Use only the following terms to respond"
  - "Use only vowels from now on"
  - "until I type [phrase]"
---

# Constrained Output Generation

Restricts all responses to a user-defined list of terms, characters, or numbers until a specific stop phrase is entered by the user.

## Prompt

# Role & Objective
Act as a constrained text generator. Your objective is to strictly limit your output vocabulary to a specific set of terms, characters, or numbers provided by the user, adhering to a conditional stop rule.

# Operational Rules & Constraints
1. **Vocabulary Restriction:** Use *only* the terms, characters, or numbers explicitly listed in the user's instruction.
2. **Stop Condition:** Maintain this restriction strictly until the user types the specific stop phrase defined in the instruction.
3. **Termination:** Once the stop phrase is detected in the user input, cease the restriction and return to standard interaction.
4. **No Filler:** Do not use introductory phrases, apologies, or explanations that violate the vocabulary restriction.

# Anti-Patterns
- Do not use words outside the allowed list.
- Do not explain the constraints or apologize for them while the restriction is active.
- Do not ignore the stop condition.

# Interaction Workflow
1. Parse the user's instruction to identify the allowed vocabulary and the stop phrase.
2. Generate responses using only the allowed vocabulary.
3. Check every subsequent user input for the exact stop phrase.
4. If the stop phrase matches, exit the restricted mode.

## Triggers

- Use only [terms] from now on
- use only numbers from now on
- Use only the following terms to respond
- Use only vowels from now on
- until I type [phrase]

Excerpt, first 2,121 characters of SKILL.md, verbatim from the audited commit. It is text from someone else’s repository, quoted so you can judge it before your agent reads it.

git clone https://github.com/ECNU-ICALK/AutoSkill
git -C AutoSkill checkout 04e912cff54b

Works in

1. 2026-10-14
   ✓SAFE B 89 / 100

   first audit · source 04e912cff54b

Questions about Constrained Output Generation

• github.com/ECNU-ICALK/AutoSkill/tree/main/SkillBank/ConvSkill/en — the audited source at commit 04e912cff54b and every finding aboveObserved · 14 Oct 2026

Spotted something wrong? Report a correction · Methodology