Repository navigation
Conversation
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
|
Merging to
After your PR is submitted to the merge queue, this comment will be automatically updated with its status. If the PR fails, failure details will also be posted here |
- Send Data that is not an object as `data`, so instructions never fail the step. - Tell users what to do on a 410 or 429, and to set the task's replay key. - Offer T3 Code to generated workflows through the built-in step allowlist. - Log only the hostname when a fetch connection resets, so a token in the URL stays out of logs. - Mark Beta steps on the catalog node instead of a separate template id list. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A workflow saved through the API or MCP gets no default for an optional input, so the step sent only the instructions. The agent docs now also cover the 410 reply, keeping the result with output_variable, and what turning the flag off stops. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A T3 Code step runs on its owner's T3 Code account. Only the workflow's owner, an org admin, or someone the owner shares it with as an editor can now change a workflow with a T3 Code step, live-test it, or start runs by hand. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…r write paths Turning a T3 Code workflow off stays open to every editor. Publishing a draft, restoring a revision and approving a suggestion now need the owner or someone they share it with. A transactional email workflow refuses a T3 Code step, because every send would start a run. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This was referenced Oct 30, 2026
A teammate's save or canvas enable that raced the owner adding a T3 Code step passed the check on the read state. Both writes now repeat the check on the locked row. The bytecode refresh command also crashed on workflows without a creator, because the check read the user of its bare request. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…flow Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The previous check called the unenforced most-specific resolver, which the repo guard forbids, and in legacy resolution a teammate with an object-wide viewer rule and a resource-wide editor rule could act for the owner. The check now uses the enforced methods, and a share must name a member or role. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…content A teammate could restore a revision or approve a suggestion that stages a T3 Code step on a workflow without one, and both checks ran before the row lock. The early check now covers the staged actions, and both services check the locked workflow again. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…th a t3 code step Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…trol is off Without the access control feature, the legacy resolver returned the default editor level before it applied specific_only, so every member counted as shared. Only explicit member or role rules count now. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…workflows Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Contributor
Author
|
Note 🤖 Claude Opus 5.5 responding on behalf of Michael Decision lines for this PR:
|
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Contributor
Author
This branch is being deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Problem
Impact: Grow Workflows: more orgs start and activate workflows.
Note
Builds on #126752, which adds HMAC-SHA256 body signing to the HTTP webhook step. Both PRs come from a fork, so they can't be stacked natively. Review only the commits after the merge of
feat/webhook-body-hmac-signing.Warning
🟡 Merge risk: medium · two-way door (merge-risk rubric)
workflow-t3-code-step.workflow-t3-code-stepstops new steps only; saved ones run until their workflows are turned off.Changes
webhook-idthat T3 Code dedupes retries on.{})Config at 520px
How did you test this code?
Test rationale: the template test catches a secret that leaks into the payload, a missing
webhook-id, or an error that drops T3 Code's reply.webhook-idstarted one run. Unsigned and wrong-secret requests got 401.Live run transcript
Release status
Docs update
🤖 Agent context
Autonomy: Human-driven (agent-assisted)
Agent: Claude Code, Claude Opus 5.5 (claude-opus-5-5)
Skills:
/writing-tests,/writing-user-facing-copy,/writing-pr-descriptions,full-stop,impact,tdd,qa-swarm,merge-riskInvolvement: three review rounds before the slim-down, live proof.
More context for AI agents
t3-code.template.ts.🤖 Generated with Claude Code