mirror of
https://github.com/openclaw/openclaw.git
synced 2026-06-06 05:51:15 +08:00
- New extensions/parallel package modeled on extensions/exa
- Wires Parallel's POST /v1/search through the generic web_search contract,
exposing Parallel's recommended {objective, search_queries} shape (plus
optional count, session_id, client_model) so the model can supply both the
natural-language goal and 2-3 short keyword queries as Parallel docs advise
- client_model lets the model report its own slug so Parallel can tailor
optimizations for the consuming model's capabilities; partitions the cache
by client_model so different models do not silently share ranked excerpts
- Honors top-level tools.web.search.{maxResults,timeoutSeconds,cacheTtlMinutes}
via the shared SDK helpers (mergeScopedSearchConfig, withTrustedWebSearchEndpoint,
buildSearchCacheKey, read/writeCachedSearchPayload)
- Auto-detect order 75; auth via PARALLEL_API_KEY or
plugins.entries.parallel.config.webSearch.apiKey
- Optional baseUrl override for proxies (e.g. Cloudflare AI Gateway)
- Threads caller-supplied session_id through follow-up calls; strips
auto-generated session_id from the shared cache to avoid cross-task leaks
- Always sends advanced_settings.max_results so result volume matches the
OpenClaw web_search default (5) instead of Parallel's default (10)
- Identifies the plugin via User-Agent header built from package version
- Runtime accepts the generic `query` arg as a fallback so the operator
CLI (openclaw capability web.search) keeps working when Parallel is the
active provider: it is promoted into the lone `search_queries` entry.
`objective` stays optional and is never synthesized from a keyword
query (Parallel documents it as natural-language intent). Agent callers
using the native objective+search_queries shape take precedence; the
schema still advertises only the native keys
- Updates the agent tool-display extractor (src/agents/tool-display-common.ts)
to recognize Parallel's objective+search_queries shape so calls render with
query context in CLI progress and Codex activity metadata
- Adds /tools/parallel-search docs page, web.md provider listing, docs nav,
labeler entry, per-plugin registration contract test, and minimal core
touch-points (legacy migrate, registration cases, providers contract list,
runtime bundled list, vitest extension paths)
632 lines
19 KiB
TypeScript
632 lines
19 KiB
TypeScript
/**
|
|
* Regression coverage for compact tool display formatting.
|
|
* Ensures tool names, actions, and details stay readable and redacted.
|
|
*/
|
|
import { describe, expect, it } from "vitest";
|
|
import { resolveToolSearchCodeDisplayTarget } from "./tool-display-common.js";
|
|
import { resolveExecDetail } from "./tool-display-exec.js";
|
|
import { formatToolDetail, formatToolSummary, resolveToolDisplay } from "./tool-display.js";
|
|
|
|
describe("tool display details", () => {
|
|
it("summarizes tool-search code targets from described tool ids", () => {
|
|
expect(
|
|
resolveToolSearchCodeDisplayTarget({
|
|
code: "const tool = await openclaw.tools.describe('openclaw:core:exec'); return await openclaw.tools.call(tool.id, { command: 'echo hi' });",
|
|
}),
|
|
).toEqual({
|
|
toolName: "openclaw:core:exec",
|
|
displayToolName: "exec",
|
|
displayArgs: { command: "echo hi" },
|
|
detail: "echo hi",
|
|
bridgeVerb: "call",
|
|
});
|
|
});
|
|
|
|
it("normalizes direct tool-search catalog ids to native display names and args", () => {
|
|
expect(
|
|
resolveToolSearchCodeDisplayTarget({
|
|
code: 'return await openclaw.tools.call("openclaw:core:exec", { command: "echo hi" });',
|
|
}),
|
|
).toEqual({
|
|
toolName: "openclaw:core:exec",
|
|
displayToolName: "exec",
|
|
displayArgs: { command: "echo hi" },
|
|
detail: "echo hi",
|
|
bridgeVerb: "call",
|
|
});
|
|
});
|
|
|
|
it("preserves JS numeric literals in tool-search call args", () => {
|
|
expect(
|
|
resolveToolSearchCodeDisplayTarget({
|
|
code: 'return await openclaw.tools.call("web_search", { query: "OpenClaw", count: 1e3, limit: +3, threshold: .5 });',
|
|
})?.displayArgs,
|
|
).toEqual({
|
|
query: "OpenClaw",
|
|
count: 1000,
|
|
limit: 3,
|
|
threshold: 0.5,
|
|
});
|
|
});
|
|
|
|
it("skips zero/false values for optional detail fields", () => {
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "sessions_spawn",
|
|
args: {
|
|
task: "double-message-bug-gpt",
|
|
label: 0,
|
|
runTimeoutSeconds: 0,
|
|
},
|
|
}),
|
|
);
|
|
|
|
expect(detail).toBe("double-message-bug-gpt");
|
|
});
|
|
|
|
it("includes only truthy boolean details", () => {
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "message",
|
|
args: {
|
|
action: "react",
|
|
provider: "discord",
|
|
to: "chan-1",
|
|
remove: false,
|
|
},
|
|
}),
|
|
);
|
|
|
|
expect(detail).toContain("provider discord");
|
|
expect(detail).toContain("to chan-1");
|
|
expect(detail).not.toContain("remove");
|
|
});
|
|
|
|
it("keeps positive numbers and true booleans", () => {
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "sessions_history",
|
|
args: {
|
|
sessionKey: "agent:main:main",
|
|
limit: 20,
|
|
includeTools: true,
|
|
},
|
|
}),
|
|
);
|
|
|
|
expect(detail).toContain("session agent:main:main");
|
|
expect(detail).toContain("limit 20");
|
|
expect(detail).toContain("tools true");
|
|
});
|
|
|
|
it("formats read/write/edit with intent-first file detail", () => {
|
|
const readDetail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "read",
|
|
args: { file_path: "/tmp/a.txt", offset: 2, limit: 2 },
|
|
}),
|
|
);
|
|
const writeDetail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "write",
|
|
args: { file_path: "/tmp/a.txt", content: "abc" },
|
|
}),
|
|
);
|
|
const editDetail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "edit",
|
|
args: { path: "/tmp/a.txt", newText: "abcd" },
|
|
}),
|
|
);
|
|
|
|
expect(readDetail).toBe("lines 2-3 from /tmp/a.txt");
|
|
expect(writeDetail).toBe("to /tmp/a.txt (3 chars)");
|
|
expect(editDetail).toBe("in /tmp/a.txt (4 chars)");
|
|
});
|
|
|
|
it("formats web_search query with quotes", () => {
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "web_search",
|
|
args: { query: "OpenClaw docs", count: 3 },
|
|
}),
|
|
);
|
|
|
|
expect(detail).toBe('for "OpenClaw docs" (top 3)');
|
|
});
|
|
|
|
it("formats web_search provider query shapes", () => {
|
|
expect(
|
|
formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "web_search",
|
|
args: { q: "Codex OAuth API key", max_results: 5 },
|
|
}),
|
|
),
|
|
).toBe('for "Codex OAuth API key" (top 5)');
|
|
|
|
expect(
|
|
formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "web_search",
|
|
args: {
|
|
search_query: [
|
|
{ q: "latest Kimi model" },
|
|
{ q: "latest Gemini model" },
|
|
{ q: "latest Claude model" },
|
|
{ q: "latest OpenAI model" },
|
|
],
|
|
},
|
|
}),
|
|
),
|
|
).toBe('for "latest Kimi model", "latest Gemini model", "latest Claude model"…');
|
|
});
|
|
|
|
it("formats Parallel's native objective + search_queries shape", () => {
|
|
expect(
|
|
formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "web_search",
|
|
args: {
|
|
objective: "Find the OpenClaw repository on GitHub",
|
|
search_queries: ["openclaw github", "openclaw repository"],
|
|
count: 5,
|
|
},
|
|
}),
|
|
),
|
|
).toBe(
|
|
'for "Find the OpenClaw repository on GitHub", "openclaw github", "openclaw repository" (top 5)',
|
|
);
|
|
});
|
|
|
|
it("summarizes exec commands with context", () => {
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "exec",
|
|
args: {
|
|
command:
|
|
"set -euo pipefail\ngit -C /Users/adityasingh/.openclaw/workspace status --short | head -n 3",
|
|
workdir: "/Users/adityasingh/.openclaw/workspace",
|
|
},
|
|
}),
|
|
);
|
|
|
|
expect(detail).toContain("check git status -> show first 3 lines");
|
|
expect(detail).toContain("(agent)");
|
|
});
|
|
|
|
it("summarizes bash commands with the same command explainer", () => {
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "bash",
|
|
args: { command: "sed -n '1,80p' extensions/discord/src/draft-stream.ts" },
|
|
detailMode: "explain",
|
|
}),
|
|
);
|
|
|
|
expect(detail).toBe("print lines 1-80 from extensions/discord/src/draft-stream.ts");
|
|
});
|
|
|
|
it("moves cd path to context suffix and appends raw command", () => {
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "exec",
|
|
args: { command: "cd ~/my-project && npm install" },
|
|
}),
|
|
);
|
|
|
|
expect(detail).toBe("install dependencies (in ~/my-project), `cd ~/my-project && npm install`");
|
|
});
|
|
|
|
it("omits raw command details in explain mode", () => {
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "exec",
|
|
args: { command: "cd ~/my-project && npm install" },
|
|
detailMode: "explain",
|
|
}),
|
|
);
|
|
|
|
expect(detail).toBe("install dependencies (in ~/my-project)");
|
|
});
|
|
|
|
it("uses compact workspace markers for common workspace paths", () => {
|
|
expect(
|
|
formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "bash",
|
|
args: { command: "git fetch", workdir: "/Users/peter/mantis-workspace/openclaw" },
|
|
detailMode: "explain",
|
|
}),
|
|
),
|
|
).toBe("fetch git changes (agent)");
|
|
|
|
expect(
|
|
formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "bash",
|
|
args: { command: "git status", workdir: "/Users/peter/Projects/openclaw" },
|
|
detailMode: "explain",
|
|
}),
|
|
),
|
|
).toBe("check git status (repo)");
|
|
|
|
expect(
|
|
formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "bash",
|
|
args: {
|
|
command: "command -v discrawl",
|
|
workdir: "/root/.openclaw/sandboxes/agent-clawsweeper-sandbox-discor-766423d0",
|
|
},
|
|
detailMode: "explain",
|
|
}),
|
|
),
|
|
).toBe("command -v discrawl");
|
|
});
|
|
|
|
it("omits bash and exec names from compact tool summaries", () => {
|
|
expect(
|
|
formatToolSummary(
|
|
resolveToolDisplay({
|
|
name: "bash",
|
|
args: { command: "git fetch", workdir: "/Users/peter/mantis-workspace/openclaw" },
|
|
detailMode: "explain",
|
|
}),
|
|
),
|
|
).toBe("🛠️ fetch git changes (agent)");
|
|
|
|
expect(
|
|
formatToolSummary(
|
|
resolveToolDisplay({
|
|
name: "web_search",
|
|
args: { query: "OpenClaw docs" },
|
|
}),
|
|
),
|
|
).toBe('🔎 Web Search: for "OpenClaw docs"');
|
|
});
|
|
|
|
it("moves cd path to context suffix with multiple stages and raw command", () => {
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "exec",
|
|
args: { command: "cd ~/my-project && npm install && npm test" },
|
|
}),
|
|
);
|
|
|
|
expect(detail).toBe(
|
|
"install dependencies → run tests (in ~/my-project), `cd ~/my-project && npm install && npm test`",
|
|
);
|
|
});
|
|
|
|
it("moves pushd path to context suffix and appends raw command", () => {
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "exec",
|
|
args: { command: "pushd /tmp && git status" },
|
|
}),
|
|
);
|
|
|
|
expect(detail).toBe("check git status (in /tmp), `pushd /tmp && git status`");
|
|
});
|
|
|
|
it("clears inferred cwd when popd is stripped from preamble", () => {
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "exec",
|
|
args: { command: "pushd /tmp && popd && npm install" },
|
|
}),
|
|
);
|
|
|
|
expect(detail).toBe("install dependencies, `pushd /tmp && popd && npm install`");
|
|
});
|
|
|
|
it("moves cd path to context suffix with || separator", () => {
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "exec",
|
|
args: { command: "cd /app || npm install" },
|
|
}),
|
|
);
|
|
|
|
// || means npm install runs when cd FAILS — cd should NOT be stripped as preamble.
|
|
// Both stages are summarized; cd is not treated as context prefix.
|
|
expect(detail).toMatch(/^run cd \/app → install dependencies/);
|
|
});
|
|
|
|
it("explicit workdir takes priority over cd path", () => {
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "exec",
|
|
args: { command: "cd /tmp && npm install", workdir: "/app" },
|
|
}),
|
|
);
|
|
|
|
expect(detail).toBe("install dependencies (in /app), `cd /tmp && npm install`");
|
|
});
|
|
|
|
it("summarizes all stages and appends raw command", () => {
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "exec",
|
|
args: { command: "git fetch && git rebase origin/main" },
|
|
}),
|
|
);
|
|
|
|
expect(detail).toBe(
|
|
"fetch git changes → rebase git branch, `git fetch && git rebase origin/main`",
|
|
);
|
|
});
|
|
|
|
it("falls back to raw command for unknown binaries", () => {
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "exec",
|
|
args: { command: "jj rebase -s abc -d main" },
|
|
}),
|
|
);
|
|
|
|
expect(detail).toBe("jj rebase -s abc -d main");
|
|
});
|
|
|
|
it("falls back to raw command for unknown binary with cwd", () => {
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "exec",
|
|
args: { command: "mycli deploy --prod", workdir: "/app" },
|
|
}),
|
|
);
|
|
|
|
expect(detail).toBe("mycli deploy --prod (in /app)");
|
|
});
|
|
|
|
it("keeps multi-stage summary when only some stages are generic", () => {
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "exec",
|
|
args: { command: "cargo build && npm test" },
|
|
}),
|
|
);
|
|
|
|
// "run cargo build" is generic, but "run tests" is known — keep joined summary
|
|
expect(detail).toMatch(/^run cargo build → run tests/);
|
|
});
|
|
|
|
it("handles standalone cd as raw command", () => {
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "exec",
|
|
args: { command: "cd /tmp" },
|
|
}),
|
|
);
|
|
|
|
// standalone cd (no following command) — treated as raw since it's generic
|
|
expect(detail).toBe("cd /tmp");
|
|
});
|
|
|
|
it("handles chained cd commands using last path", () => {
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "exec",
|
|
args: { command: "cd /tmp && cd /app" },
|
|
}),
|
|
);
|
|
|
|
// both cd's are preamble; last path wins
|
|
expect(detail).toBe("cd /tmp && cd /app (in /app)");
|
|
});
|
|
|
|
it("respects quotes when splitting preamble separators", () => {
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "exec",
|
|
args: { command: 'export MSG="foo && bar" && echo test' },
|
|
}),
|
|
);
|
|
|
|
// The && inside quotes must not be treated as a separator —
|
|
// summary line should be "print text", not "run export" (which would happen
|
|
// if the quoted && was mistaken for a real separator).
|
|
expect(detail).toMatch(/^print text/);
|
|
});
|
|
|
|
it("recognizes heredoc/inline script exec details", () => {
|
|
const pyDetail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "exec",
|
|
args: {
|
|
command: "python3 <<PY\nprint('x')\nPY",
|
|
workdir: "/Users/adityasingh/.openclaw/workspace",
|
|
},
|
|
}),
|
|
);
|
|
const nodeCheckDetail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "exec",
|
|
args: {
|
|
command: "node --check /tmp/test.js",
|
|
workdir: "/Users/adityasingh/.openclaw/workspace",
|
|
},
|
|
}),
|
|
);
|
|
const nodeShortCheckDetail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "exec",
|
|
args: {
|
|
command: "node -c /tmp/test.js",
|
|
workdir: "/Users/adityasingh/.openclaw/workspace",
|
|
},
|
|
}),
|
|
);
|
|
|
|
expect(pyDetail).toContain("run python3 inline script (heredoc)");
|
|
expect(nodeCheckDetail).toContain("check js syntax for /tmp/test.js");
|
|
expect(nodeShortCheckDetail).toContain("check js syntax for /tmp/test.js");
|
|
});
|
|
|
|
it("appends node name to exec detail when node is set", () => {
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "exec",
|
|
args: {
|
|
command: "docker pull pihole/pihole:latest",
|
|
host: "node",
|
|
node: "raspberrypi",
|
|
},
|
|
}),
|
|
);
|
|
|
|
expect(detail).toContain("node: raspberrypi");
|
|
});
|
|
|
|
it("includes both cwd and node name in exec detail for known commands", () => {
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "exec",
|
|
args: {
|
|
command: "npm install",
|
|
workdir: "/app",
|
|
host: "node",
|
|
node: "raspberrypi",
|
|
},
|
|
}),
|
|
);
|
|
|
|
expect(detail).toContain("(in /app)");
|
|
expect(detail).toContain("node: raspberrypi");
|
|
});
|
|
|
|
it("omits node label when node param is absent or empty", () => {
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "exec",
|
|
args: { command: "npm install", host: "gateway" },
|
|
}),
|
|
);
|
|
|
|
expect(detail).not.toContain("node:");
|
|
});
|
|
|
|
it("omits node label when host is not 'node' even if node is set", () => {
|
|
for (const host of ["gateway", "sandbox", "auto"]) {
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "exec",
|
|
args: { command: "npm install", host, node: "raspberrypi" },
|
|
}),
|
|
);
|
|
|
|
expect(detail).not.toContain("node:");
|
|
}
|
|
});
|
|
});
|
|
|
|
describe("compactRawCommand middle truncation", () => {
|
|
it("preserves start and end of long commands", () => {
|
|
// Use an unknown binary so resolveExecDetail returns the compact raw form directly.
|
|
const longCommand =
|
|
"/opt/custom/bin/my-processor --input /data/warehouse/2024/q1/transactions/raw/batch_001.csv --output /data/warehouse/2024/q1/transactions/processed/batch_001_clean.csv";
|
|
const result = resolveExecDetail({ command: longCommand });
|
|
// Should contain the start of the command
|
|
expect(result).toContain("/opt/custom/bin/my-processor");
|
|
// Should contain the end (filename)
|
|
expect(result).toContain("batch_001_clean.csv");
|
|
// Should contain the ellipsis for middle truncation
|
|
expect(result).toContain("…");
|
|
// Ellipsis should be in the middle, not at the end
|
|
expect(result).not.toMatch(/…$/);
|
|
});
|
|
|
|
it("does not truncate short commands", () => {
|
|
// Use an unknown binary so resolveExecDetail returns the compact raw form directly.
|
|
const result = resolveExecDetail({ command: "/opt/custom/bin/my-tool --version" });
|
|
expect(result).toBe("/opt/custom/bin/my-tool --version");
|
|
});
|
|
|
|
it("redacts credential-like tails before middle truncation", () => {
|
|
// The --token flag and its value sit in the middle of a long command.
|
|
// Without redaction-before-truncation, middle truncation could cut out
|
|
// the --token flag context but preserve the raw secret at the tail.
|
|
const longCommand =
|
|
"/opt/custom/bin/deploy --region us-east-1 --token sk-proj-ABCDEFGHIJKLMNOP1234567890abcdefghij --output /data/results/deploy-output.json";
|
|
const result = resolveExecDetail({ command: longCommand });
|
|
// The sk- prefixed token must be redacted (masked) before truncation
|
|
expect(result).not.toContain("ABCDEFGHIJKLMNOP1234567890abcdefghij");
|
|
});
|
|
|
|
it("uses the canonical tool payload redactor before compacting raw commands", () => {
|
|
const longCommand =
|
|
"/opt/custom/bin/deploy --aws-key AKIDABCDEFGHIJKLMNOP1234567890 --output /data/results/deploy-output.json";
|
|
const result = resolveExecDetail({ command: longCommand });
|
|
|
|
expect(result).not.toContain("AKIDABCDEFGHIJKLMNOP1234567890");
|
|
expect(result).toContain("AKIDAB…7890");
|
|
});
|
|
});
|
|
|
|
describe("coerceDisplayValue middle truncation", () => {
|
|
it("preserves start and end of long string values", () => {
|
|
const longPath =
|
|
"/usr/local/share/very/deeply/nested/directory/structure/" +
|
|
"a".repeat(150) +
|
|
"/important-file.txt";
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "sessions_spawn",
|
|
args: { task: longPath },
|
|
}),
|
|
);
|
|
// Should contain the start of the path
|
|
expect(detail).toContain("/usr/local/share/");
|
|
// Should contain the end (filename)
|
|
expect(detail).toContain("important-file.txt");
|
|
// Should contain the ellipsis for middle truncation
|
|
expect(detail).toContain("…");
|
|
});
|
|
|
|
it("does not truncate short string values", () => {
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "sessions_spawn",
|
|
args: { task: "short-task-name" },
|
|
}),
|
|
);
|
|
expect(detail).toBe("short-task-name");
|
|
expect(detail).not.toContain("…");
|
|
});
|
|
|
|
it("redacts credential-like values in long generic string details", () => {
|
|
// A long string whose tail contains a GitHub PAT. Without
|
|
// redaction-before-truncation, middle truncation could preserve
|
|
// the raw token at the tail after its prefix context is cut.
|
|
const longValue =
|
|
"Deploying service to production cluster with auth ghp_ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnop and " +
|
|
"x".repeat(200) +
|
|
" final-step";
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "sessions_spawn",
|
|
args: { task: longValue },
|
|
}),
|
|
);
|
|
// The ghp_ token must be redacted before truncation
|
|
expect(detail).not.toContain("ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnop");
|
|
});
|
|
|
|
it("uses the canonical tool payload redactor before compacting string details", () => {
|
|
const longValue =
|
|
"Deploying with AWS key AKIDABCDEFGHIJKLMNOP1234567890 and " +
|
|
"x".repeat(200) +
|
|
" final-step";
|
|
const detail = formatToolDetail(
|
|
resolveToolDisplay({
|
|
name: "sessions_spawn",
|
|
args: { task: longValue },
|
|
}),
|
|
);
|
|
|
|
expect(detail).not.toContain("AKIDABCDEFGHIJKLMNOP1234567890");
|
|
expect(detail).toContain("AKIDAB…7890");
|
|
});
|
|
});
|