fix(agents): split system prompt cache prefix by transport by vincentkoc · Pull Request #59054 · openclaw/openclaw

vincentkoc · 2026-04-01T14:03:20Z

Summary

Problem: Anthropic-family prompt caching still lost KV reuse when dynamic lab/session/system additions changed, because the current transport stack no longer had the old cache seam and the shared payload policy cache-tagged the whole system block.
Why it matters: with many labs and large prompt surfaces, per-turn dynamic suffix churn rewrites expensive stable prompt prefixes and multiplies cache misses.
What changed: restored an internal system-prompt cache boundary as a shared helper, moved context-engine prompt additions behind that boundary, split Anthropic system blocks into cached stable and uncached dynamic regions in the shared payload policy, and stripped the internal marker before emission across non-Anthropic transports plus CLI backend system-prompt args.
What did NOT change (scope boundary): provider-visible prompt text is unchanged apart from removing the internal marker; no config/env changes were introduced; no new cache writes happen when cacheRetention: "none".

Change Type (select all)

Scope (select all touched areas)

Linked Issue/PR

Closes [Feature]: Multi-Block Prompt Caching for Anthropic Models #50511
Closes cache_control not applied to system prompt on direct Anthropic provider path (cacheRead=0) #57958
Related fix(anthropic): split system prompt into static/dynamic blocks for stable cache prefix #53225
This PR fixes a bug or regression

Root Cause / Regression History (if applicable)

Root cause: the original fix lived in an older wrapper path, but upstream provider transport/KV work centralized Anthropic request shaping in shared payload policy code and removed the effective cache seam from the current path.
Secondary regressions: the initial refresh restored the Anthropic seam but missed the OpenAI Completions serializer path, and the CLI backend path still forwarded the internal cache boundary marker unchanged via systemPromptArg.
Missing detection / guardrail: there was no current-main regression test ensuring dynamic system additions, including context-engine prepends, stay outside the cached Anthropic prefix while the internal marker is stripped for all other emission paths.
Contributing context (if known): upstream merged a large amount of KV/provider transport work after the original PR branch was cut, so the old patch shape no longer matched the live request path.

Regression Test Plan (if applicable)

Coverage level that should have caught this:
- Unit test
- Seam / integration test
- End-to-end test
- Existing coverage already sufficient
Target test or file:
- src/agents/system-prompt-cache-boundary.test.ts
- src/agents/anthropic-payload-policy.test.ts
- src/agents/openai-transport-stream.test.ts
- src/agents/openai-ws-stream.test.ts
- src/agents/cli-runner.helpers.test.ts
Scenario the test should lock in: Anthropic-family requests cache only the stable system prefix; dynamic lab/session suffix content stays uncached; OpenAI Responses, OpenAI Completions, OpenAI WebSocket, Google transport paths, and CLI backend system-prompt args strip the internal boundary marker before emission.
Why this is the smallest reliable guardrail: the regression lives in prompt assembly plus transport/CLI payload shaping, not in end-user workflow code.
Existing test that already covers this (if any): prompt stability coverage existed, but it did not protect the transport-layer cache seam.
If no new test is added, why not: N/A.

User-visible / Behavior Changes

Anthropic-family sessions can preserve the stable system prompt prefix across turns again even when lab/group/session additions change later in the prompt. Other request paths keep seeing the same prompt text, with the internal boundary stripped before emission.

Diagram (if applicable)

Before:
stable prompt prefix + dynamic lab/session suffix -> one cached Anthropic system block -> suffix churn rewrites the full prefix
non-Anthropic/CLI emission paths -> some paths still forwarded the raw internal boundary marker

After:
stable prompt prefix -> cached Anthropic block
dynamic lab/session suffix -> uncached Anthropic block
non-Anthropic/CLI emission paths -> same prompt text with boundary stripped before emission

Security Impact (required)

New permissions/capabilities? (No)
Secrets/tokens handling changed? (No)
New/changed network calls? (No)
Command/tool execution surface changed? (No)
Data access scope changed? (No)
If any Yes, explain risk + mitigation: N/A

Repro + Verification

Environment

OS: macOS
Runtime/container: Node 25 / pnpm
Model/provider: Anthropic-family payload shaping plus OpenAI/Google/WebSocket/CLI strip paths
Integration/channel (if any): N/A
Relevant config (redacted): prompt boundary present vs absent; cacheRetention long vs none

Steps

Build a system prompt containing the internal cache boundary.
Apply Anthropic payload policy with cache retention enabled.
Apply non-Anthropic transport request builders or CLI backend arg builders using the same prompt.

Expected

Anthropic-family payloads split cached stable system content from uncached dynamic suffix content.
Other emission paths strip the internal boundary marker and emit unchanged prompt text semantics.

Actual

Before this update, current-main shaping cache-tagged the whole Anthropic system block and had no equivalent seam for later dynamic additions.
Before the follow-up fixes in this PR, the OpenAI Completions path and the CLI backend systemPromptArg path could still emit the raw internal marker.

Evidence

Attach at least one:

Failing test/log before + passing after
Trace/log snippets
Screenshot/recording
Perf numbers (if relevant)

Human Verification (required)

Verified scenarios: git diff --check; direct module import sanity for touched source files; direct sanity check that OpenAI Completions now strips the internal boundary marker.
Edge cases checked: boundary stripped when cache retention is disabled; context-engine prepend path moves additions behind the cache seam; non-Anthropic request builders strip the marker before emission, including OpenAI Completions; CLI backend system-prompt args also strip the marker before emission.
What you did not verify: full Vitest lanes, pnpm check, pnpm build, or live provider telemetry in this update path.

Review Conversations

I replied to or resolved every bot review conversation I addressed in this PR.
I left unresolved only the conversations that still need reviewer or maintainer judgment.

If a bot review conversation is addressed by this PR, resolve that conversation yourself. Do not leave bot review conversation cleanup for maintainers.

Compatibility / Migration

Backward compatible? (Yes)
Config/env changes? (No)
Migration needed? (No)
If yes, exact upgrade steps: N/A

Risks and Mitigations

Risk: an internal boundary marker could leak into provider requests on an emission path that was missed.
- Mitigation: Anthropic strips/splits in shared payload policy; OpenAI Responses, OpenAI Completions, OpenAI WebSocket, Google builders, and CLI backend system-prompt args explicitly strip before emission.
Risk: moving the seam too early would reduce cache leverage for large stable bootstrap context.
- Mitigation: the boundary is placed late in buildAgentSystemPrompt() so stable prompt context stays cacheable and only dynamic lab/session additions sit behind it.

vincentkoc · 2026-04-01T14:03:28Z

Supersedes #53225. Refreshed onto current main, revalidated locally, and kept the change scoped to Anthropic/Bedrock system-prompt cache splitting with boundary stripping on all paths.

martingarramon

Clean approach — splitting at a known boundary is much better than trying to diff system prompt sections across turns.

One question on the hot path: the wrapper is applied unconditionally to ctx.agent.streamFn for all providers, so every request (OpenAI, Ollama, etc.) runs through the onPayload callback to strip the marker even when splitAndCache is false. The per-request cost is low (one .includes() per system block), but this is the streaming hot path. Was there a reason not to gate the wrapper application behind something like system prompt actually contains the marker? Or is the simplicity of always-apply worth the negligible overhead?

Also noticed the test covers the array-system split path and the string-system strip path, but not the case where splitAndCache is true with a plain-string system prompt (the typeof system === "string" branch at line 59 of the new file). That path has a deliberate design choice — falling back to strip instead of split — which seems worth a test to document the intent.

greptile-apps · 2026-04-04T03:12:40Z

Greptile Summary

This PR restores an Anthropic-family system prompt cache seam by introducing a new SYSTEM_PROMPT_CACHE_BOUNDARY marker, inserting it in buildAgentSystemPrompt, splitting Anthropic payloads into a cached stable block and an uncached dynamic block, and stripping the marker before emission on Google, OpenAI Responses, and WebSocket paths.

P1 – boundary leaks into completions path: buildOpenAICompletionsParams passes the raw context to the external convertMessages from @mariozechner/pi-ai/openai-completions without first stripping the marker (compare line 226 in the same file where convertResponsesMessages explicitly calls stripSystemPromptCacheBoundary). Every Completions-API provider (Mistral, Moonshot, xAI-completions, etc.) will receive `\

\

` embedded in the system message. The PR description states OpenAI builders explicitly strip before emission, but the completions builder is the missed path.

Confidence Score: 4/5

Safe to review but has one active defect: the OpenAI Completions path leaks the internal cache boundary marker into provider requests, contradicting the PR's stated mitigation.

The Anthropic split logic, Google strip, OpenAI Responses strip, and WebSocket strip are all correctly implemented and tested. One transport path — OpenAI Completions (buildOpenAICompletionsParams) — passes the raw context including the boundary marker to an external library without pre-stripping, causing the marker to appear in system messages for all Completions-API providers. This is a present defect in the changed code, not a speculative future risk, so it warrants a P1 and reduces the score to 4.

src/agents/openai-transport-stream.ts — specifically buildOpenAICompletionsParams (around line 1232) needs a stripSystemPromptCacheBoundary call before passing context to the external convertMessages, and a matching test in openai-transport-stream.test.ts.

Comments Outside Diff (1)

src/agents/openai-transport-stream.ts, line 1232 (link)

Boundary marker leaks into completions path

convertMessages from @mariozechner/pi-ai/openai-completions receives context with context.systemPrompt still containing the raw  marker. Unlike convertResponsesMessages (line 226) which calls stripSystemPromptCacheBoundary(context.systemPrompt) before building the messages, this external function has no awareness of the internal marker, so it will embed the literal \n\n string in the system role message sent to every OpenAI Completions provider (Mistral, Moonshot, xAI-completions, etc.). The PR description states OpenAI builders explicitly strip before emission, but the completions builder is the missed path.

A straightforward fix is to strip before constructing params and pass the cleaned context, or strip inside buildOpenAICompletionsParams itself before calling convertMessages.

Prompt To Fix With AI

This is a comment left during a code review.
Path: src/agents/openai-transport-stream.ts
Line: 1232

Comment:
**Boundary marker leaks into completions path**

`convertMessages` from `@mariozechner/pi-ai/openai-completions` receives `context` with `context.systemPrompt` still containing the raw `<!-- OPENCLAW_CACHE_BOUNDARY -->` marker. Unlike `convertResponsesMessages` (line 226) which calls `stripSystemPromptCacheBoundary(context.systemPrompt)` before building the messages, this external function has no awareness of the internal marker, so it will embed the literal `\n<!-- OPENCLAW_CACHE_BOUNDARY -->\n` string in the `system` role message sent to every OpenAI Completions provider (Mistral, Moonshot, xAI-completions, etc.). The PR description states OpenAI builders explicitly strip before emission, but the completions builder is the missed path.

A straightforward fix is to strip before constructing params and pass the cleaned context, or strip inside `buildOpenAICompletionsParams` itself before calling `convertMessages`.

How can I resolve this? If you propose a fix, please make it concise.

Prompt To Fix All With AI

This is a comment left during a code review.
Path: src/agents/openai-transport-stream.ts
Line: 1232

Comment:
**Boundary marker leaks into completions path**

`convertMessages` from `@mariozechner/pi-ai/openai-completions` receives `context` with `context.systemPrompt` still containing the raw `<!-- OPENCLAW_CACHE_BOUNDARY -->` marker. Unlike `convertResponsesMessages` (line 226) which calls `stripSystemPromptCacheBoundary(context.systemPrompt)` before building the messages, this external function has no awareness of the internal marker, so it will embed the literal `\n<!-- OPENCLAW_CACHE_BOUNDARY -->\n` string in the `system` role message sent to every OpenAI Completions provider (Mistral, Moonshot, xAI-completions, etc.). The PR description states OpenAI builders explicitly strip before emission, but the completions builder is the missed path.

A straightforward fix is to strip before constructing params and pass the cleaned context, or strip inside `buildOpenAICompletionsParams` itself before calling `convertMessages`.

How can I resolve this? If you propose a fix, please make it concise.

---

This is a comment left during a code review.
Path: src/agents/openai-transport-stream.test.ts
Line: 443-466

Comment:
**No boundary-stripping test for the completions path**

The existing test covers `buildOpenAIResponsesParams` stripping the cache boundary, but there is no equivalent test for `buildOpenAICompletionsParams`. This is the coverage gap that allows the leak described above to go undetected; adding a parallel test case for `buildOpenAICompletionsParams` with a boundary-containing `systemPrompt` would lock in the fix once the stripping logic is added.

How can I resolve this? If you propose a fix, please make it concise.

_{Reviews (1): Last reviewed commit: "Merge branch 'main' into vk/pr-53225-cac..." | Re-trigger Greptile}

greptile-apps · 2026-04-04T03:12:47Z

+  it("strips the internal cache boundary from OpenAI system prompts", () => {
+    const params = buildOpenAIResponsesParams(
+      {
+        id: "gpt-5.4",
+        name: "GPT-5.4",
+        api: "openai-responses",
+        provider: "openai",
+        baseUrl: "https://api.openai.com/v1",
+        reasoning: true,
+        input: ["text"],
+        cost: { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 },
+        contextWindow: 200000,
+        maxTokens: 8192,
+      } satisfies Model<"openai-responses">,
+      {
+        systemPrompt: `Stable prefix${SYSTEM_PROMPT_CACHE_BOUNDARY}Dynamic suffix`,
+        messages: [],
+        tools: [],
+      } as never,
+      undefined,
+    ) as { input?: Array<{ content?: string }> };
+
+    expect(params.input?.[0]?.content).toBe("Stable prefix\nDynamic suffix");
+  });


No boundary-stripping test for the completions path

The existing test covers buildOpenAIResponsesParams stripping the cache boundary, but there is no equivalent test for buildOpenAICompletionsParams. This is the coverage gap that allows the leak described above to go undetected; adding a parallel test case for buildOpenAICompletionsParams with a boundary-containing systemPrompt would lock in the fix once the stripping logic is added.

Prompt To Fix With AI

This is a comment left during a code review. Path: src/agents/openai-transport-stream.test.ts Line: 443-466 Comment: **No boundary-stripping test for the completions path** The existing test covers `buildOpenAIResponsesParams` stripping the cache boundary, but there is no equivalent test for `buildOpenAICompletionsParams`. This is the coverage gap that allows the leak described above to go undetected; adding a parallel test case for `buildOpenAICompletionsParams` with a boundary-containing `systemPrompt` would lock in the fix once the stripping logic is added. How can I resolve this? If you propose a fix, please make it concise.

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: b5dd1fee90

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: c90fd94a64

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector · 2026-04-04T03:31:04Z

+  // Keep large stable prompt context above this seam so Anthropic-family
+  // transports can reuse it across labs and turns. Dynamic group/session
+  // additions below it are the primary cache invalidators.
+  lines.push(SYSTEM_PROMPT_CACHE_BOUNDARY);


Route default streamSimple calls through boundary stripping

buildAgentSystemPrompt now inserts SYSTEM_PROMPT_CACHE_BOUNDARY unconditionally, but the marker is only removed in transport-specific builders. For standard sessions without request.proxy/request.tls, registerProviderStreamForModel does not install those transport wrappers (src/agents/provider-transport-stream.ts), and resolveEmbeddedAgentStreamFn falls back to currentStreamFn/streamSimple (src/agents/pi-embedded-runner/run/attempt.ts:243-257), so provider requests still receive the raw  text. This is a prompt-content regression on default provider paths and also skips the Anthropic stable/dynamic cache split outside the custom transport flow.

Useful? React with 👍 / 👎.

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 2614d0fc0c

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector · 2026-04-04T03:44:38Z

+  // Keep large stable prompt context above this seam so Anthropic-family
+  // transports can reuse it across labs and turns. Dynamic group/session
+  // additions below it are the primary cache invalidators.
+  lines.push(SYSTEM_PROMPT_CACHE_BOUNDARY);


Strip cache boundary before provider-owned stream calls

buildAgentSystemPrompt now unconditionally inserts SYSTEM_PROMPT_CACHE_BOUNDARY, but provider-owned createStreamFn paths are not routed through the new boundary-stripping transports. For example, registerProviderStreamForModel can return the Ollama plugin stream (extensions/ollama/index.ts), and that stream forwards context.systemPrompt directly into the outbound system message (extensions/ollama/src/stream.ts), so requests will now include the internal  marker in provider-visible prompt text. This is a prompt-content regression for bundled/custom provider plugins that do not implement their own stripping.

Useful? React with 👍 / 👎.

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 7340201cc0

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector · 2026-04-04T04:37:52Z

+      context.systemPrompt
+        ? {
+            ...context,
+            systemPrompt: stripSystemPromptCacheBoundary(context.systemPrompt),
+          }


Preserve Anthropic cache seam in provider stream wrapper

resolveEmbeddedAgentStreamFn now strips SYSTEM_PROMPT_CACHE_BOUNDARY for every providerStreamFn call, but Anthropic proxy/TLS models reach this branch via registerProviderStreamForModel and rely on applyAnthropicPayloadPolicyToParams to split cached stable system text from dynamic suffix text. Stripping first removes that seam, so those Anthropic requests revert to cache-tagging the whole system block and lose prefix cache reuse whenever dynamic additions change.

Useful? React with 👍 / 👎.

chatgpt-codex-connector · 2026-04-04T04:37:52Z

+  if (!isTransportAwareApiSupported(model.api)) {
+    return undefined;


Handle boundary stripping for non-transport-aware APIs

createBoundaryAwareStreamFnForModel returns undefined for APIs outside SUPPORTED_TRANSPORT_APIS, so those sessions fall back to streamSimple without boundary cleanup. Since the resolver still accepts APIs like bedrock-converse-stream, github-copilot, and openai-codex-responses (src/agents/pi-embedded-runner/model.ts:103-114), those providers now receive the internal  marker in prompt text.

Useful? React with 👍 / 👎.

…59054) * fix(agents): restore Anthropic prompt cache seam * fix(agents): strip cache boundary for completions * fix(agents): strip cache boundary for cli backends * chore(changelog): note cross-transport cache boundary rollout * fix(agents): route default stream fallbacks through boundary shapers * fix(agents): strip cache boundary for provider streams

vincentkoc self-assigned this Apr 1, 2026

vincentkoc mentioned this pull request Apr 1, 2026

fix(anthropic): split system prompt into static/dynamic blocks for stable cache prefix #53225

Closed

9 tasks

openclaw-barnacle Bot added agents Agent runtime and tooling size: S maintainer Maintainer-authored PR labels Apr 1, 2026

martingarramon reviewed Apr 2, 2026

View reviewed changes

xinbenlv mentioned this pull request Apr 3, 2026

Suggestion: strengthen contributor-first PR workflow and credit practices #60108

Closed

vincentkoc force-pushed the vk/pr-53225-cache-split branch from acf2f84 to f3be414 Compare April 4, 2026 03:06

openclaw-barnacle Bot added size: M and removed size: S labels Apr 4, 2026

vincentkoc marked this pull request as ready for review April 4, 2026 03:08

greptile-apps Bot reviewed Apr 4, 2026

View reviewed changes

chatgpt-codex-connector Bot reviewed Apr 4, 2026

View reviewed changes

Comment thread src/agents/openai-transport-stream.ts

Comment thread src/agents/system-prompt.ts

vincentkoc changed the title ~~fix(agents): split Anthropic system prompt cache prefix~~ fix(agents): split system prompt cache prefix by transport Apr 4, 2026

chatgpt-codex-connector Bot reviewed Apr 4, 2026

View reviewed changes

vincentkoc added 5 commits April 4, 2026 12:38

fix(agents): restore Anthropic prompt cache seam

703df6d

fix(agents): strip cache boundary for completions

e7e6114

fix(agents): strip cache boundary for cli backends

e49171c

chore(changelog): note cross-transport cache boundary rollout

7b96e8d

fix(agents): route default stream fallbacks through boundary shapers

2614d0f

vincentkoc force-pushed the vk/pr-53225-cache-split branch from 93e191d to 2614d0f Compare April 4, 2026 03:39

chatgpt-codex-connector Bot reviewed Apr 4, 2026

View reviewed changes

This comment has been minimized.

Sign in to view

fix(agents): strip cache boundary for provider streams

293b582

openclaw-barnacle Bot added size: L and removed size: M labels Apr 4, 2026

Merge branch 'main' into vk/pr-53225-cache-split

7340201

vincentkoc merged commit 64f2890 into main Apr 4, 2026
7 checks passed

vincentkoc deleted the vk/pr-53225-cache-split branch April 4, 2026 04:32

chatgpt-codex-connector Bot reviewed Apr 4, 2026

View reviewed changes

vincentkoc mentioned this pull request Apr 5, 2026

fix(cache): order stable project context before heartbeat #61236

Merged

25 tasks

100yenadmin mentioned this pull request Apr 7, 2026

feat: prompt cache keep-warm pings to prevent TTL expiry between turns #62475

Open

clawsweeper Bot mentioned this pull request Apr 30, 2026

[Bug]: Anthropic prompt cache busted on every turn: per-message metadata injected inside system block with cache_control #75300

Closed

		if (!isTransportAwareApiSupported(model.api)) {
		return undefined;

Uh oh!

Conversation

vincentkoc commented Apr 1, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Change Type (select all)

Scope (select all touched areas)

Linked Issue/PR

Root Cause / Regression History (if applicable)

Regression Test Plan (if applicable)

User-visible / Behavior Changes

Diagram (if applicable)

Security Impact (required)

Repro + Verification

Environment

Steps

Expected

Actual

Evidence

Human Verification (required)

Review Conversations

Compatibility / Migration

Risks and Mitigations

Uh oh!

vincentkoc commented Apr 1, 2026

Uh oh!

martingarramon left a comment

Choose a reason for hiding this comment

Uh oh!

greptile-apps Bot commented Apr 4, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Greptile Summary

Confidence Score: 4/5

Comments Outside Diff (1)

Uh oh!

greptile-apps Bot Apr 4, 2026

Choose a reason for hiding this comment

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

Uh oh!

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

chatgpt-codex-connector Bot Apr 4, 2026

Choose a reason for hiding this comment

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

chatgpt-codex-connector Bot Apr 4, 2026

Choose a reason for hiding this comment

Uh oh!

This comment has been minimized.

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

chatgpt-codex-connector Bot Apr 4, 2026

Choose a reason for hiding this comment

Uh oh!

chatgpt-codex-connector Bot Apr 4, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

vincentkoc commented Apr 1, 2026 •

edited

Loading

greptile-apps Bot commented Apr 4, 2026 •

edited

Loading