v2 message format and upgrade to ai sdk v5 (#743)

Co-authored-by: GitHub Action <action@github.com>
Co-authored-by: Liang-Shih Lin <liangshihlin@proton.me>
Co-authored-by: Dominik Engelhardt <dominikengelhardt@ymail.com>
Co-authored-by: Jay V <air@live.ca>
Co-authored-by: adamdottv <2363879+adamdottv@users.noreply.github.com>
This commit is contained in:
Dax
2025-07-07 15:53:43 -04:00
committed by GitHub
parent 76b2e4539c
commit f884766445
116 changed files with 4707 additions and 6950 deletions

View File

@@ -16,14 +16,8 @@ export const ReadTool = Tool.define({
description: DESCRIPTION,
parameters: z.object({
filePath: z.string().describe("The path to the file to read"),
offset: z
.number()
.describe("The line number to start reading from (0-based)")
.optional(),
limit: z
.number()
.describe("The number of lines to read (defaults to 2000)")
.optional(),
offset: z.number().describe("The line number to start reading from (0-based)").optional(),
limit: z.number().describe("The number of lines to read (defaults to 2000)").optional(),
}),
async execute(params, ctx) {
let filePath = params.filePath
@@ -40,16 +34,13 @@ export const ReadTool = Tool.define({
const suggestions = dirEntries
.filter(
(entry) =>
entry.toLowerCase().includes(base.toLowerCase()) ||
base.toLowerCase().includes(entry.toLowerCase()),
entry.toLowerCase().includes(base.toLowerCase()) || base.toLowerCase().includes(entry.toLowerCase()),
)
.map((entry) => path.join(dir, entry))
.slice(0, 3)
if (suggestions.length > 0) {
throw new Error(
`File not found: ${filePath}\n\nDid you mean one of these?\n${suggestions.join("\n")}`,
)
throw new Error(`File not found: ${filePath}\n\nDid you mean one of these?\n${suggestions.join("\n")}`)
}
throw new Error(`File not found: ${filePath}`)
@@ -57,21 +48,14 @@ export const ReadTool = Tool.define({
const stats = await file.stat()
if (stats.size > MAX_READ_SIZE)
throw new Error(
`File is too large (${stats.size} bytes). Maximum size is ${MAX_READ_SIZE} bytes`,
)
throw new Error(`File is too large (${stats.size} bytes). Maximum size is ${MAX_READ_SIZE} bytes`)
const limit = params.limit ?? DEFAULT_READ_LIMIT
const offset = params.offset || 0
const isImage = isImageFile(filePath)
if (isImage)
throw new Error(
`This is an image file of type: ${isImage}\nUse a different tool to process images`,
)
if (isImage) throw new Error(`This is an image file of type: ${isImage}\nUse a different tool to process images`)
const lines = await file.text().then((text) => text.split("\n"))
const raw = lines.slice(offset, offset + limit).map((line) => {
return line.length > MAX_LINE_LENGTH
? line.substring(0, MAX_LINE_LENGTH) + "..."
: line
return line.length > MAX_LINE_LENGTH ? line.substring(0, MAX_LINE_LENGTH) + "..." : line
})
const content = raw.map((line, index) => {
return `${(index + offset + 1).toString().padStart(5, "0")}| ${line}`
@@ -82,9 +66,7 @@ export const ReadTool = Tool.define({
output += content.join("\n")
if (lines.length > offset + content.length) {
output += `\n\n(File has more lines. Use 'offset' parameter to read beyond line ${
offset + content.length
})`
output += `\n\n(File has more lines. Use 'offset' parameter to read beyond line ${offset + content.length})`
}
output += "\n</file>"
@@ -93,10 +75,10 @@ export const ReadTool = Tool.define({
FileTime.read(ctx.sessionID, filePath)
return {
title: path.relative(App.info().path.root, filePath),
output,
metadata: {
preview,
title: path.relative(App.info().path.root, filePath),
},
}
},