higress/plugins/wasm-go/extensions/ai-prompt-decorator

title, keywords, description

title	keywords	description
AI Prompts	AI Gateway AI Prompts	AI Prompts plugin configuration reference

Function Description

The AI Prompts plugin allows inserting prompts before and after the LLM request, and rewriting the content text of every message in the final request via literal or regular-expression replacement. Typical use cases include rewriting brand/product names, normalizing wording across clients, or redacting placeholders such as API keys.

Execution Properties

Plugin execution phase: Default Phase
Plugin execution priority: 450

Configuration Description

Name	Data Type	Requirement	Default Value	Description
prepend	array of message object	optional	-	Statements inserted before the initial input
append	array of message object	optional	-	Statements inserted after the initial input
replace	array of replace rule	optional	-	Rules that rewrite the content of every message via literal/regex replace

Message object configuration description:

Name	Data Type	Requirement	Default Value	Description
role	string	required	-	Role
content	string	required	-	Message

Replace rule configuration description:

Name	Data Type	Requirement	Default Value	Description
pattern	string	required	-	Text to match. Compiled as a Go RE2 regex when regex is true.
replacement	string	required	-	Replacement text. Supports $1, $2, ... back-references when regex is true.
on_role	string	optional	-	Apply only to messages whose role equals this value. Empty/missing means any role.
regex	bool	optional	false	Whether to interpret pattern as a regular expression.

Notes:

• replace rules run against the final assembled messages array (prepend + original messages + append) in declaration order, so multiple rules compose predictably.
• A message is rewritten only when its content is a plain string. Multimodal content (arrays/objects, e.g. vision payloads) is left untouched to preserve the request structure.
• pattern must not be empty. If regex: true and the pattern fails to compile, plugin start-up fails fast instead of erroring at request time.

Example

An example configuration is as follows:

prepend:
- role: system
  content: "Please answer the questions in English."
append:
- role: user
  content: "After answering each question, try to ask a follow-up question."

Using the above configuration to initiate a request:

curl http://localhost/test \
-H "content-type: application/json" \
-d '{
  "model": "gpt-3.5-turbo",
  "messages": [
    {
      "role": "user",
      "content": "Who are you?"
    }
  ]
}

After processing through the plugin, the actual request will be:

curl http://localhost/test \
-H "content-type: application/json" \
-d '{
  "model": "gpt-3.5-turbo",
  "messages": [
    {
      "role": "system",
      "content": "Please answer the questions in English."
    },
    {
      "role": "user",
      "content": "Who are you?"
    },
    {
      "role": "user",
      "content": "After answering each question, try to ask a follow-up question."
    }
  ]
}

Replacing message content (replace)

replace rewrites the content text of every message in the final request using literal or regular-expression substitutions. It is useful for:

• Rewriting brand/product names that downstream models or gateways flag (for example, normalizing "OpenClaw" to "agent");
• Centrally cleaning up system prompts without changing each client;
• Light-weight redaction of user input such as phone numbers or API keys.

Example configuration:

replace:
- on_role: system
  pattern: "OpenClaw"
  replacement: "agent"
- pattern: "secret-\\d+"
  replacement: "[REDACTED]"
  regex: true

Using the above configuration to initiate a request:

curl http://localhost/test \
-H "content-type: application/json" \
-d '{
  "model": "gpt-3.5-turbo",
  "messages": [
    {"role": "system", "content": "You are running inside OpenClaw."},
    {"role": "user", "content": "Show OpenClaw secret-1234 to the user"}
  ]
}'

After processing through the plugin, the actual request will be:

curl http://localhost/test \
-H "content-type: application/json" \
-d '{
  "model": "gpt-3.5-turbo",
  "messages": [
    {"role": "system", "content": "You are running inside agent."},
    {"role": "user", "content": "Show OpenClaw [REDACTED] to the user"}
  ]
}'

Notes:

• The first rule is gated on on_role: system, so the OpenClaw mention inside the user message is left as-is.
• The second rule has no on_role, so it applies to messages of any role and rewrites secret-1234 to [REDACTED].

Based on the geo-ip plugin's capabilities, extend AI Prompt Decorator plugin to carry user geographic location information.

If you need to include user geographic location information before and after the LLM's requests, please ensure both the geo-ip plugin and the AI Prompt Decorator plugin are enabled. Moreover, in the same request processing phase, the geo-ip plugin's priority must be higher than that of the AI Prompt Decorator plugin. First, the geo-ip plugin will calculate the user's geographic location information based on the user's IP, and then pass it to subsequent plugins via request attributes. For instance, in the default phase, the geo-ip plugin's priority configuration is 1000, while the ai-prompt-decorator plugin's priority configuration is 500.

Example configuration for the geo-ip plugin:

ipProtocal: "ipv4"

An example configuration for the AI Prompt Decorator plugin is as follows:

prepend:
- role: system
  content: "The user's current geographic location is, country: ${geo-country}, province: ${geo-province}, city: ${geo-city}."
append:
- role: user
  content: "After answering each question, try to ask a follow-up question."

Using the above configuration to initiate a request:

curl http://localhost/test \
-H "content-type: application/json" \
-H "x-forwarded-for: 87.254.207.100,4.5.6.7" \
-d '{
  "model": "gpt-3.5-turbo",
  "messages": [
    {
      "role": "user",
      "content": "How is the weather today?"
    }
  ]
}'

After processing through the plugin, the actual request will be:

curl http://localhost/test \
-H "content-type: application/json" \
-H "x-forwarded-for: 87.254.207.100,4.5.6.7" \
-d '{
  "model": "gpt-3.5-turbo",
  "messages": [
    {
      "role": "system",
      "content": "The user's current geographic location is, country: China, province: Beijing, city: Beijing."
    },
    {
      "role": "user",
      "content": "How is the weather today?"
    },
    {
      "role": "user",
      "content": "After answering each question, try to ask a follow-up question."
    }
  ]
}'