feat: migrate agents to @relayfile/relay-helpers#20
Conversation
Replaces the hardcoded VFS-path writes with the ergonomic, catalog-backed
clients from @relayfile/relay-helpers — paths now come from the adapter catalog
instead of string literals in each agent.
- slack post/dm -> slackClient().post(channel,text) / .dm(user,text)
(hn-monitor, vendor-monitor, spotify-releases, review, repo-hygiene)
- linear get/comment -> linearClient().getIssue(id) / .comment(id,body) (linear)
- linear create+comment needing the writeback receipt -> relayClient('linear')
.write('issues'|'comments', ...) (granola)
- github comment / merge -> githubClient().comment(...) / .mergePullRequest(...)
(review, repo-hygiene)
Reads that aren't catalog writeback resources stay on the raw runtime VFS
helpers: github pulls/meta.json + notion journal (repo-hygiene), linear teams
(granola). typecheck clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
|
CodeAnt AI is reviewing your PR. |
📝 WalkthroughWalkthroughThis PR refactors all agent handlers to replace VFS-based integration writes with direct client calls from ChangesRelay helpers client migration
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~20 minutes Possibly related PRs
Poem
🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
codeant-ai
Bot
added
the
size:M
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request refactors several agents to use helper clients from @relayfile/relay-helpers (such as slackClient, linearClient, githubClient, and relayClient) instead of manually constructing paths and calling writeJsonFile or readJsonFile from @agentworkforce/runtime. Feedback on the changes highlights a potential runtime crash in linear/agent.ts if the issue is not found, suggesting a defensive check before processing the retrieved issue.
Important
The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.
| 'getIssue', | ||
| `/linear/issues/${encodeSegment(issueId)}.json` | ||
| ); | ||
| const issue = await linear.getIssue<LinearIssue>(issueId); |
There was a problem hiding this comment.
To prevent potential runtime crashes, we should defensively check if the issue was successfully retrieved before passing it to parseRepo. If the issue is not found or the API call returns undefined/null, calling parseRepo(issue) will throw a TypeError when attempting to access issue.title.
| const issue = await linear.getIssue<LinearIssue>(issueId); | |
| const issue = await linear.getIssue<LinearIssue>(issueId); | |
| if (!issue) { | |
| ctx.log('warn', 'linear-implementer.issue-not-found', { issueId }); | |
| return; | |
| } |
There was a problem hiding this comment.
Declining: linearClient().getIssue<LinearIssue>(issueId) returns a non-nullable LinearIssue — on a missing/unreadable file it throws (readJsonFile → RelayfileWritebackError), it never resolves to undefined/null. So if (!issue) is dead code, and the behavior is unchanged from the pre-migration readJsonFile<LinearIssue>(...) which threw the same way.
|
CodeAnt AI finished reviewing your PR. |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@review/agent.ts`:
- Around line 98-99: Ensure GitHub and Slack writes are idempotent by adding a
dedupe guard before calling githubClient().comment(...) and
slackClient().post(...): generate a stable idempotency key per PR/run (e.g.,
using pr.owner, pr.repo, pr.number and the review run id) and either pass it to
the client if it supports idempotency headers or check-and-skip by querying
existing PR comments / Slack messages for that key; apply the same guard in
failReviewRun() and mergePr() so they first verify whether the comment/post with
that idempotency key already exists before posting to avoid duplicate writes on
retries.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro Plus
Run ID: 04e1d687-d95f-47ce-a994-781a84210528
⛔ Files ignored due to path filters (1)
package-lock.jsonis excluded by!**/package-lock.json
📒 Files selected for processing (9)
.gitignoregranola/agent.tshn-monitor/agent.tslinear/agent.tspackage.jsonrepo-hygiene/agent.tsreview/agent.tsspotify-releases/agent.tsvendor-monitor/agent.ts
| await githubClient().comment({ owner: pr.owner, repo: pr.repo, number: pr.number }, body); | ||
| } |
There was a problem hiding this comment.
🧩 Analysis chain
🌐 Web query:
Does @relayfile/relay-helpers githubClient/slackClient provide idempotency keys or dedupe for comment/post calls?
💡 Result:
No evidence found that @relayfile/relay-helpers’ githubClient/slackClient supports idempotency keys or request deduplication for comment/post (write) calls. In the searches performed, nothing in the relay-helpers GitHub code references idempotency-key headers, idempotency-key parameters, or any dedupe mechanism for comment/post operations; results only surfaced unrelated “relay” projects and general idempotency-key articles, not the @relayfile/relay-helpers implementation itself [1][2][3]. Therefore, based on the sources retrieved, the correct conclusion is that idempotency/dedupe is not provided (or at least not discoverable/documented) in githubClient/slackClient for comment/post calls. If you share the relay-helpers repo link (or the specific file paths for githubClient/slackClient), I can re-run a more targeted source-code search for “idempotency”, “dedupe”, “Idempotency-Key”, or “X-Request-ID” within those exact implementations.
Citations:
- 1: https://github.com/AgentWorkforce/relayfile-providers
- 2: https://dev.to/apikumo/idempotency-keys-how-to-make-your-api-calls-safe-to-retry-2fln
- 3: https://dev.to/apikumo/idempotency-keys-the-api-pattern-that-saves-you-from-duplicate-payments-and-phantom-records-51b2
🏁 Script executed:
#!/bin/bash
set -euo pipefail
echo "== Repo file tree (focus) =="
ls -R
echo "== Find references to relay-helpers / githubClient / slackClient =="
rg -n "relay-helpers|githubClient\(|slackClient\(|failReviewRun|post\(|comment\(" -S .
echo "== Show review/agent.ts around referenced lines =="
sed -n '70,180p' review/agent.ts
echo "== Locate package.json/lock to identify where `@relayfile/relay-helpers` comes from =="
ls
fd -a "package.json" .
fd -a "package-lock.json" .
fd -a "pnpm-lock.yaml" .
fd -a "yarn.lock" .Repository: AgentWorkforce/agents
Length of output: 8940
🏁 Script executed:
#!/bin/bash
set -euo pipefail
# Check if node_modules exists (so we can inspect actual relay-helpers code)
if [ -d "node_modules/@relayfile/relay-helpers" ]; then
echo "Found local node_modules/@relayfile/relay-helpers"
ls -R node_modules/@relayfile/relay-helpers | head -n 200
echo "== Search for idempotency/dedupe markers in relay-helpers =="
rg -n "Idempotency|idempotency|dedupe|dedup|X-Request-ID|requestId|retry|deduplication" node_modules/@relayfile/relay-helpers -S || true
echo "== Search for GitHub comment/post call wrappers =="
rg -n "comment\\(|slackClient\\(\\)|post\\(|mergePullRequest\\(" node_modules/@relayfile/relay-helpers -S || true
else
echo "node_modules/@relayfile/relay-helpers not present in this checkout."
fi
echo "== Search for retry/at-least-once hints in review harness entrypoints =="
rg -n "retry|retries|redeliver|at-?least-?once|idempotent|dedup" review/agent.ts . -S || true
echo "== Show any local docs about idempotency for relay-helpers in repo =="
rg -n "`@relayfile/relay-helpers`|relay-helpers" -S README.md review/README.md . || trueRepository: AgentWorkforce/agents
Length of output: 1197
🌐 Web query:
@relayfile/relay-helpers githubClient slackClient source githubClient().comment slackClient().post idempotency dedupe
💡 Result:
I couldn’t find any public, authoritative evidence for the specific internal chain you referenced: githubClient().comment slackClient().post idempotency dedupe in @relayfile/relay-helpers. What I did find from the public Relayfile ecosystem sources is: 1) Relayfile has a JavaScript SDK and a client (RelayFileClient) and the SDK “adds X-Correlation-Id automatically for API calls” (useful for tracing/replay debugging, but this is not the same as idempotent write deduplication). [1] 2) Relayfile adapters exist for GitHub and Slack (so there are likely code paths where GitHub comments and Slack posts happen), but the sources surfaced in search results didn’t include the @relayfile/relay-helpers implementation details about how dedupe/idempotency is handled for GitHub comment creation or Slack message posting. [2] 3) Some other open-source codebases that integrate with Slack/GitHub implement message/event deduplication explicitly (TTL caches keyed by message/event id), showing the common approach when providers/clients can retry deliveries (e.g., a TTL-based deduplicator). This indicates the pattern you probably want, but it is not proof of Relayfile’s behavior. [3] If you can share the exact GitHub URL (or paste the relevant code) for @relayfile/relay-helpers and where githubClient() / slackClient() are defined, I can re-research specifically for how dedupe keys are generated and where they are applied (e.g., per-post hash, per-operation idempotency keys, or provider-side headers).
Citations:
- 1: https://npmx.dev/package/@relayfile/sdk
- 2: https://github.com/AgentWorkforce/relayfile-adapters
- 3: https://github.com/NousResearch/hermes-agent/blob/main/gateway/platforms/helpers.py
🌐 Web query:
site:github.com "@relayfile/relay-helpers" githubClient comment implementation
💡 Result:
I couldn’t find any authoritative public source that shows the actual “@relayfile/relay-helpers githubClient comment” implementation (i.e., the code for how @relayfile/relay-helpers uses a GitHub client to create/update PR comments). What I was able to verify from public GitHub sources is that the relayfile ecosystem uses adapters/providers to write back to GitHub comments via an adapter path like /github/repos///pulls/<pull_number>/comments, and that the adapter consumes a provider for auth. Example usage pattern from the relayfile providers repo: - It constructs a RelayFileClient with a Relayfile token. - It uses a GitHubAdapter from @relayfile/adapter-github. - It calls adapter.writeback(...) with a path pointing to a PR comments endpoint and a payload like { body: "Looks good!" }. This demonstrates where a “comment implementation” would be invoked, but not the internal code inside @relayfile/relay-helpers itself. [1] If you can share the specific function/class name inside @relayfile/relay-helpers (or a link to the repository/package source), I can re-run targeted searches for that exact file and produce a fully code-grounded answer for the githubClient comment implementation.
Citations:
🌐 Web query:
site:npmjs.com "@relayfile/relay-helpers" idempotency slackClient
💡 Result:
I couldn’t find any authoritative or directly relevant information about the package name “@relayfile/relay-helpers” (including any mention of “idempotency” or “slackClient”). The web results I found around “relay-helpers” appear to be unrelated packages (e.g., Relay GraphQL helpers/network-layer, or other “relay” libraries), not “@relayfile/relay-helpers”. So, I can’t confidently answer how (or whether) “@relayfile/relay-helpers” implements idempotency for a “slackClient” from evidence. If you paste the repository link (GitHub URL) or the relevant file path(s) (e.g., where slackClient/idempotency are implemented), I can re-run the research specifically on that code/docs and give a grounded answer.
Citations:
- 1: https://www.npmjs.com/package/relay-network-layer
- 2: https://www.npmjs.com/package/react-relay-network-layer
- 3: https://www.npmjs.com/package/relay-hooks
- 4: https://www.npmjs.com/package/@datafire/slack
- 5: https://www.npmjs.com/package/@solana-developers/helpers
Make GitHub/Slack write operations idempotent under handler retries.
review/agent.ts performs unconditional direct write calls; if the platform redelivers/retries the event, these calls will be re-issued, with no dedupe/idempotency guard at this layer.
- Success path:
githubClient().comment(...)(lines 98-99) andslackClient().post(...)(lines 104-109). - Failure path:
failReviewRun()always re-posts both the GitHub comment and Slack warning (lines 125-132) before throwing. - Merge path:
mergePr()always posts the Slack “Merged PR” message (line 151).
No accessible evidence shows @relayfile/relay-helpers githubClient() / slackClient() adding Idempotency-Key/request deduplication for comment/post operations.
await githubClient().comment({ owner: pr.owner, repo: pr.repo, number: pr.number }, body);Require idempotent behavior (e.g., provider/client-side idempotency keys + retry-safe semantics, or local/per-PR-per-run dedupe) or ensure these handler paths cannot be retried/redelivered.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@review/agent.ts` around lines 98 - 99, Ensure GitHub and Slack writes are
idempotent by adding a dedupe guard before calling githubClient().comment(...)
and slackClient().post(...): generate a stable idempotency key per PR/run (e.g.,
using pr.owner, pr.repo, pr.number and the review run id) and either pass it to
the client if it supports idempotency headers or check-and-skip by querying
existing PR comments / Slack messages for that key; apply the same guard in
failReviewRun() and mergePr() so they first verify whether the comment/post with
that idempotency key already exists before posting to avoid duplicate writes on
retries.
![cubic-dev-ai[bot]](https://avatars.githubusercontent.com/in/1082092?s=60&v=4){kind=small}
There was a problem hiding this comment.
2 issues found across 10 files
Prompt for AI agents (unresolved issues)
Check if these issues are valid — if so, understand the root cause of each and fix them. If appropriate, use sub-agents to investigate and fix each issue separately.
<file name="review/agent.ts">
<violation number="1" location="review/agent.ts:98">
P2: Guard this outbound comment write with idempotency/deduplication. On handler redelivery, this path can post duplicate PR comments.</violation>
<violation number="2" location="review/agent.ts:146">
P1: Merge confirmation uses a truthiness check, which can falsely accept non-boolean values like `'false'` as a successful merge.</violation>
</file>
Reply with feedback, questions, or to request a fix.
Re-trigger cubic
| if (merged !== true && merged !== 'true') { | ||
| // mergePullRequest surfaces the writeback worker's merge outcome as `merged`. | ||
| // A false/unconfirmed result means we shouldn't pretend the merge landed. | ||
| if (!result.merged) { |
There was a problem hiding this comment.
P1: Merge confirmation uses a truthiness check, which can falsely accept non-boolean values like 'false' as a successful merge.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At review/agent.ts, line 146:
<comment>Merge confirmation uses a truthiness check, which can falsely accept non-boolean values like `'false'` as a successful merge.</comment>
<file context>
@@ -140,36 +122,33 @@ async function failReviewRun(ctx: WorkforceCtx, pr: Pr, reason: string): Promise
- if (merged !== true && merged !== 'true') {
+ // mergePullRequest surfaces the writeback worker's merge outcome as `merged`.
+ // A false/unconfirmed result means we shouldn't pretend the merge landed.
+ if (!result.merged) {
throw new Error(`GitHub did not confirm PR #${pr.number} in ${pr.owner}/${pr.repo} was merged.`);
}
</file context>
There was a problem hiding this comment.
Declining: result.merged here is the normalized boolean returned by githubClient().mergePullRequest(...) (Promise<{ merged: boolean; sha?: string }>), not the raw receipt. The client already collapses receipt.merged (which may be true/'true'/false/'false'/undefined-with-sha) into a real boolean — 'false' → false — so if (!result.merged) correctly rejects a failed merge. The string case the old code guarded (merged !== true && merged !== 'true') is handled inside the client now.
There was a problem hiding this comment.
Thanks for the feedback! I've saved this as a new learning to improve future reviews.
| const client = vfsClient(); | ||
| if (body) { | ||
| await writeJsonFile(client, 'github', 'comment', prCommentPath(pr), { body }); | ||
| await githubClient().comment({ owner: pr.owner, repo: pr.repo, number: pr.number }, body); |
There was a problem hiding this comment.
P2: Guard this outbound comment write with idempotency/deduplication. On handler redelivery, this path can post duplicate PR comments.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At review/agent.ts, line 98:
<comment>Guard this outbound comment write with idempotency/deduplication. On handler redelivery, this path can post duplicate PR comments.</comment>
<file context>
@@ -112,19 +94,19 @@ async function reviewAndFix(ctx: WorkforceCtx, pr: Pr): Promise<void> {
- const client = vfsClient();
if (body) {
- await writeJsonFile(client, 'github', 'comment', prCommentPath(pr), { body });
+ await githubClient().comment({ owner: pr.owner, repo: pr.repo, number: pr.number }, body);
}
</file context>
There was a problem hiding this comment.
Acknowledged, but deferring out of this PR: this is a mechanical migration (same writes, just via relay-helpers) — the duplicate-on-redelivery behavior is pre-existing, not introduced here. Also the runtime dedupes redeliveries on event.id (WorkforceEventBase.id — "the runtime dedupes on this"), so the handler shouldn't re-run for the same event. Full write-idempotency (stable keys / querying existing comments) is a heavy lift and relay-helpers/the writeback worker don't expose idempotency headers today — better tracked as its own issue than bundled into the migration.
1 participant
User description
Migrates all agents off the hardcoded VFS-path writes to the ergonomic, catalog-backed clients from
@relayfile/relay-helpers(0.3.29). Paths now resolve from the adapter catalog instead of string literals duplicated in each agent — so a provider path change inrelayfile-adaptersflows through automatically.writeJsonFile(c,'slack','post','/slack/channels/…',…)slackClient().post(channel, text)slackClient().dm(user, text)readJsonFile/writeJsonFilelinear pathslinearClient().getIssue(id)/.comment(id, body)relayClient('linear').write('issues'|'comments', …)githubClient().comment(…)/.mergePullRequest(…)+slackClient().post(…)githubClient().comment(…)+slackClient().post(…)Reads that aren't catalog writeback resources stay on the raw runtime VFS helpers (correct — they're not in the catalog): github
pulls/meta.json+ the Notion journal (repo-hygiene), the linear teams listing (granola).Verified the github merge payload contract while migrating: the adapter reads
object.method ?? object.merge_method(and snake/camel for commit fields), somergePullRequestis compatible.npm run typecheckclean.🤖 Generated with Claude Code
Summary by cubic
Migrate all agents from hardcoded VFS write paths to catalog-backed clients in
@relayfile/relay-helpersfor cleaner APIs and automatic adapter path updates. Slack, Linear, and GitHub writebacks now use typed helpers; non-catalog reads are unchanged.Refactors
slackClient().post(channel, text)/.dm(user, text)inhn-monitor,vendor-monitor,spotify-releases,review,repo-hygiene.linearClient().getIssue(id)/.comment(id, body);granolausesrelayClient('linear').write('issues'|'comments', …)to keep writeback receipts.githubClient().comment(...)and.mergePullRequest(...)inreviewandrepo-hygiene(merge response validated viamerged).pulls/meta.json, Notion journal, Linear teams).Dependencies
@relayfile/relay-helpers@^0.3.29(pulls@relayfile/adapter-core@^0.3.29)..gitignoreadds.claude/;npm run typecheckclean.Written for commit 62d77a1. Summary will update on new commits.
CodeAnt-AI Description
Move agent Slack, GitHub, and Linear actions to relay helper clients
What Changed
Impact
✅ Fewer broken integration paths✅ More reliable Slack, GitHub, and Linear actions✅ Safer PR merge confirmation💡 Usage Guide
Checking Your Pull Request
Every time you make a pull request, our system automatically looks through it. We check for security issues, mistakes in how you're setting up your infrastructure, and common code problems. We do this to make sure your changes are solid and won't cause any trouble later.
Talking to CodeAnt AI
Got a question or need a hand with something in your pull request? You can easily get in touch with CodeAnt AI right here. Just type the following in a comment on your pull request, and replace "Your question here" with whatever you want to ask:
This lets you have a chat with CodeAnt AI about your pull request, making it easier to understand and improve your code.
Example
Preserve Org Learnings with CodeAnt
You can record team preferences so CodeAnt AI applies them in future reviews. Reply directly to the specific CodeAnt AI suggestion (in the same thread) and replace "Your feedback here" with your input:
This helps CodeAnt AI learn and adapt to your team's coding style and standards.
Example
Retrigger review
Ask CodeAnt AI to review the PR again, by typing:
Check Your Repository Health
To analyze the health of your code repository, visit our dashboard at https://app.codeant.ai. This tool helps you identify potential issues and areas for improvement in your codebase, ensuring your repository maintains high standards of code health.