mirror of
https://github.com/AmruthPillai/Reactive-Resume.git
synced 2026-06-22 04:11:55 +10:00
Amruth Pillai
6d8d8f6e55
* chore(ai): remove local AI store now that providers live server-side
The Zustand-based useAIStore has been replaced by the server-side
aiProviders oRPC router (encrypted credentials persisted in DB).
Delete the dead store + tests, drop the ./store export, and remove
zustand/immer deps which are no longer referenced anywhere in
packages/ai/src/.
* feat(agent): archive/delete actions and read-only state for agent threads
- Backend: mark archived threads as read-only in threads.get and reject
messages.send with CONFLICT when the thread is archived.
- Frontend: render archived threads in the sidebar with muted styling and
an Archived badge; add a per-thread dropdown menu in the chat header
with Archive (non-destructive) and Delete (with confirmation); show a
read-only banner above the message list that disambiguates archived
vs. missing-resource causes; suppress the Retry and Stop buttons in
read-only mode.
- Tests: new packages/api/src/services/agent.test.ts covering the
archived-thread isReadOnly flag and the archived-thread send refusal.
* fix(agent): abort run on archive and verify ownership before deleting thread
- threads.archive: before flipping status, abort any in-flight run controller
and clear the active-run state on the thread; cleanup failures are logged
but do not block the status update.
- threads.delete: assert thread ownership via getThread before destructive
work so an authenticated user cannot wipe another user's attachment rows
by passing a foreign threadId.
Adds focused tests for both behaviors.
* feat(agent): display patch diffs and surface revert conflicts
Render apply_resume_patch tool messages with a status-aware card (applied/
reverted/conflicted), expandable operation list, and a Revert button that
correctly handles RESUME_VERSION_CONFLICT responses. Adds unit tests for
the inverse-patch builder and the agentService.actions.revert flow.
* chore(agent): remove out-of-scope attachment tests accidentally added in Task 6
The Task 6 commit (73ef1acca) accidentally re-introduced three attachment-
related tests that belong to a separate task:
- `buildAttachmentModelParts > converts text, image, supported binary, and
unsupported attachments into model parts`
- `agentService.messages.send > persists the user message with file UI parts
and links selected attachments to it` (was failing — the `ToolLoopAgent`
mock is not callable as a constructor)
- `agentService.messages.send > rejects attachments that are missing, foreign,
or already linked before persisting a message`
These were likely re-added during a stash recovery and were not requested
for Task 6, whose scope was limited to the `agentService.actions.revert`
flow. Remove them along with the helpers/fixtures (`buildAttachment`,
`buildActiveThread`, `selectWhereResult`, `selectOrderByResult`) that they
were the only consumers of. `selectLimitResult` is preserved because it is
used by the revert tests.
* chore(agent): configure runtime dependencies
* feat(db): add agent workspace schema
* feat(api): add agent backend services
* feat(web): add agent workspace UI
* chore(agent): remove legacy builder assistant
* test(agent): make agent stream mocks constructible
* chore(web): remove unused resume replacement hook
* feat(api): add unsafe AI base URL flag
* chore(dev): expose local services in compose
* fix(web): normalize resume preview gaps
* feat(api): improve agent tool handling
* feat(web): polish agent workspace UI
* chore: update dependencies
* fix(api,web): address PR review feedback for agent workspace
Security/correctness:
- Restrict AI provider URLs to http/https even in unsafe mode
- Stop exposing Redis on host network by default
- Make .env.local optional and drop app profile in compose.dev.yml
- Store agent attachments with private ACL on S3
- Reset provider test status when provider/model/baseURL changes
- Decouple non-agent AI endpoints from REDIS_URL requirement
- Fix JSON Patch add inverse for existing object members
- Wrap resume patch + agent action insert in db transaction
- Validate partialMessage at runtime and rate-limit attachment uploads
- Add unique index on agent_messages (thread_id, sequence)
UX/bugs:
- Mark agent thread route as ssr: false and guard SSE chunk parsing
- Show config-specific banner only on known configuration error
- Gate AI provider checks behind loading state in resume import
- Fix relative-time formatter blank gap between 45-59 seconds
- Clarify thread delete confirmation message
Polish:
- Raise ENCRYPTION_SECRET minimum to 32 characters
- Bucket AI rate limits by resumeId/threadId/messageId
- Trim form values before submitting AI provider config
- Use single key identifier and nullish-coalesce baseURL display
* fix: address ai agent review feedback
* fix: preserve mobile agent chat state
* docs: add ai agent workspace guides
* feat: introduce design system for Reactive Resume
84 lines
4.3 KiB
Plaintext
84 lines
4.3 KiB
Plaintext
---
|
|
title: "AI Agent Tools"
|
|
description: "Understand the tools available to the AI Agent workspace and how they affect resume drafts."
|
|
---
|
|
|
|
The AI Agent workspace can use a curated set of tools while it chats with you. You do not call these tools directly; the agent chooses them when your request needs resume data, web context, attachments, questions, or a resume patch.
|
|
|
|
## Tool activity in chat
|
|
|
|
Tool activity appears inside the conversation. Some activity is collapsed by default so the chat stays readable.
|
|
|
|
Applied resume patches are shown as a small inline **Patch applied** item. Open it to inspect the raw JSON payload.
|
|
|
|
<Frame caption="Patch details and tool activity in the AI Agent chat">
|
|
<img
|
|
src="/images/guides/ai-agent-tools/screenshot-1.webp"
|
|
alt="AI Agent chat showing an applied patch with raw JSON details"
|
|
/>
|
|
</Frame>
|
|
|
|
## Available tools
|
|
|
|
| Tool | What it does | Example request |
|
|
| --- | --- | --- |
|
|
| `read_resume` | Reads the current AI draft and gives the agent the resume data it can safely edit. | "What are the weakest parts of this resume?" |
|
|
| `fetch_url` | Fetches public HTTPS pages, extracts readable content, and returns it as agent input. | "Tailor this resume to this job description: `https://example.com/job`" |
|
|
| Provider-native search | Uses the selected provider's native web search when that provider/model supports it. | "Research this company and adjust the summary for its product area." |
|
|
| `read_attachment` | Reads extracted text from attached plain text, Markdown, or JSON files. Other supported attachments, such as images or PDFs, are passed to the model when the selected provider can use them. | "Use the attached notes to update the keywords." |
|
|
| `ask_user_question` | Shows a question card with answer choices when the agent needs your decision. | "Ask me before changing the career narrative." |
|
|
| `apply_resume_patch` | Applies a JSON Patch to the AI draft and stores enough history to revert it later. | "Change the visible name to Amruth Pillai." |
|
|
|
|
## Resume patches
|
|
|
|
Resume patches are rooted at the resume data object. For example, the visible resume name is patched at `/basics/name`.
|
|
|
|
When a patch is applied:
|
|
|
|
- the AI draft updates immediately;
|
|
- the raw JSON Patch is available from the **Patch applied** details;
|
|
- an inverse patch is stored for revert;
|
|
- the resume preview refreshes to show the updated draft.
|
|
|
|
If the resume changed after the patch was generated, revert or apply can fail with a version conflict. In that case, ask the agent to retry from the latest draft.
|
|
|
|
## Web access
|
|
|
|
The agent can fetch public HTTPS URLs so it can read job descriptions, company pages, or other public context you provide.
|
|
|
|
For self-hosted deployments:
|
|
|
|
- private, loopback, and non-HTTPS URLs are blocked by default;
|
|
- Cloudflare URL extraction is optional and is used only as a fallback when local readability extraction fails;
|
|
- unsafe/private AI provider base URLs require `FLAG_ALLOW_UNSAFE_AI_BASE_URL=true`, which should only be used on trusted self-hosted deployments.
|
|
|
|
## Attachments
|
|
|
|
Attach files from the chat composer when the agent needs extra context, such as:
|
|
|
|
- a job description PDF;
|
|
- a portfolio brief;
|
|
- a screenshot;
|
|
- a plain text note with constraints.
|
|
|
|
Self-hosted deployments need S3-compatible storage for private agent attachments. Local filesystem storage rejects private objects.
|
|
|
|
## Good prompts for tool use
|
|
|
|
Use direct prompts that tell the agent what context to use and how cautious to be:
|
|
|
|
- "Fetch this role URL, identify the most important keywords, and apply a conservative patch."
|
|
- "Read the attached job description and ask me before changing anything outside the summary."
|
|
- "Compare my current projects against this company page and suggest only truthful wording."
|
|
- "Apply a patch for the visible resume name, then show me what changed."
|
|
|
|
## When tools are unavailable
|
|
|
|
Tool use can be limited by the selected provider, deployment configuration, or thread state.
|
|
|
|
- A deleted provider makes the thread read-only; a disabled or untested provider blocks new agent runs until it is enabled and tested again.
|
|
- A deleted working resume makes the thread read-only.
|
|
- Archived threads cannot receive new messages.
|
|
- URL fetching may fail for private URLs, blocked hosts, non-HTML pages, or pages that cannot be extracted.
|
|
- Attachments may fail if private object storage is not configured.
|