refactor: drop the data-qa agent, make analyst the data-question agent - #1239
Conversation
…agent #1217 added an opt-in `data-qa` agent registered behind `ALTIMATE_DATA_QA_PROFILE` (prompt = builder-minus-dbt-packs, full read/write builder permissions). Drop it: - The name is misleading — "QA" reads as quality-assurance in a data tool, not question-answering. - It overlaps the existing `analyst` agent ("Read-only data exploration and analysis. Cannot modify files or run destructive SQL.") while keeping FULL read/write builder permissions — a slim-prompt agent that can run destructive SQL, the exact safety concern that got PR #1215 closed. - `analyst` already IS the safe read-only data agent; `analyst.txt` is already lean (no dbt-build/finish-protocol weight), so the insight behind data-qa (dropping build packs is score-neutral on read-only data questions) requires no prompt change — analyst already embodies it. Changes: - `agent/agent.ts`: remove the conditional `data-qa` registration block (env flag, config-entry, and `default_agent` opt-ins) and the now-unused `Flag` import. - `altimate/prompts/profiles.ts`: remove `DATA_QA_PROFILE` and `PROMPT_DATA_QA`. `BUILDER_PROFILE` / `PROMPT_BUILDER` / `assemble` / `FRAGMENTS` — the pack-split architecture from #1217 — are untouched; the builder byte-identity pin (sha256 `17663410dd9a…`) stays green. - `session/termination.ts`: drop `data-qa` from `COMPLETION_CONTRACT_AGENTS` (builder stays). - Delete `test/agent/data-qa-profile.test.ts`; remove the "data-qa profile composition" describe block from `test/altimate/prompt-profiles.test.ts` (builder byte-identity + determinism tests kept); drop the data-qa assertions from `test/session/termination.test.ts`. - Docs: no page ever documented `data-qa`, so nothing to remove there. Updated the agent-reference docs (`configure/agents.md`, `data-engineering/agent-modes.md`, `getting-started.md`, `getting-started/quickstart.md`) to explicitly frame `analyst` as the agent for asking questions about your data. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VqnuBDGkh1ZT65Ti7e6DHZ
![claude[bot]](https://avatars.githubusercontent.com/in/1236702?s=60&v=4){kind=small}
There was a problem hiding this comment.
Claude Code Review
This repository is configured for manual code reviews. Comment @claude review for a one-time review, or @claude review always to subscribe this PR to a review on every future push.
Tip: disable this comment in your organization's Code Review settings.
Bugbot couldn't run - usage limit reachedBugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit. A user or team admin can review and increase usage limits in the Cursor dashboard. (requestId: serverGenReqId_8f76a9b1-b361-488a-93be-cadccf2a56fd) |
|
👋 This PR was automatically closed by our quality checks. Common reasons:
If you believe this was a mistake, please open an issue explaining your intended contribution and a maintainer will help you. |
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
full receipts (1 session)
builder ·
|
|
👋 This PR was automatically closed by our quality checks. Common reasons:
If you believe this was a mistake, please open an issue explaining your intended contribution and a maintainer will help you. |
1 similar comment
|
👋 This PR was automatically closed by our quality checks. Common reasons:
If you believe this was a mistake, please open an issue explaining your intended contribution and a maintainer will help you. |
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Repository UI Review profile: CHILL Plan: Team Run ID: 📒 Files selected for processing (2)
Included review availability: Your plan provides up to 4 included reviews per hour; 3 remain after this review. 📝 WalkthroughWalkthroughThe change revises Analyst documentation, migrates stale native ChangesAnalyst and data-qa updates
Estimated code review effort: 2 (Simple) | ~15 minutes Merge Risk: ⚪ Minimal · up to Persisted native data-qa defaults now fall back to the read-only analyst agent while explicitly configured data-qa agents remain unchanged. The documented migration paths are covered, and no merge-blocking risk remains. Poem
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✨ 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 |
|
👋 This PR was automatically closed by our quality checks. Common reasons:
If you believe this was a mistake, please open an issue explaining your intended contribution and a maintainer will help you. |
![kilo-code-bot[bot]](https://avatars.githubusercontent.com/in/2193792?s=60&v=4){kind=small}
Code Review SummaryStatus: No Issues Found | Recommendation: Merge Files Reviewed (2 files)
Previous Review Summaries (2 snapshots, latest commit f61cd61)Current summary above is authoritative. Previous snapshots are kept for context only. Previous review (commit f61cd61)Status: No Issues Found | Recommendation: Merge Files Reviewed (3 files)
Previous review (commit 3576a54)Status: 1 Issue Found | Recommendation: Address before merge Overview
Issue Details (click to expand)SUGGESTION
Files Reviewed (11 files)
Reviewed by deepseek-v4-pro · Input: 57.6K · Output: 11K · Cached: 615.6K Review guidance: REVIEW.md from base branch |
|
👋 This PR was automatically closed by our quality checks. Common reasons:
If you believe this was a mistake, please open an issue explaining your intended contribution and a maintainer will help you. |
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 3576a54d33
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".
![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 11 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="packages/opencode/src/session/termination.ts">
<violation number="1" location="packages/opencode/src/session/termination.ts:186">
P3: This comment incorrectly claims builder was the only previous carrier, but the previous set also injected the instruction into `data-qa`. Describe the builder path as unchanged and state that the removed `data-qa` path no longer receives it.</violation>
</file>
<file name="packages/opencode/src/agent/agent.ts">
<violation number="1" location="packages/opencode/src/agent/agent.ts:334">
P1: When a legacy `agent.data-qa` entry remains, the generic custom-agent path recreates it as a mode-`all` agent with the default permissive rules. Reject this legacy key or migrate it to `analyst` before generic custom-agent creation so SQL writes retain the intended protection.</violation>
</file>
Reply with feedback, questions, or to request a fix.
Re-trigger cubic
- codex/cubic (P1): verified and locked in with a regression test that a
leftover `agent: {"data-qa": {...}}` config entry — a user upgrading
from #1217 who never cleaned it out — resolves safely through the
existing generic custom-agent path (already exercised by "custom agent
from config creates new agent"): mode "all", `native: false`, no
resurrected `data-qa` prompt or permissions, no crash. Same generic
path already covers a stale `default_agent: "data-qa"` with no
matching `agent.data-qa` entry — it throws the same clear, tested
"default agent \"data-qa\" not found" error any other stale/renamed
`default_agent` value produces (see "defaultAgent throws when
default_agent points to non-existent agent"). No migration code
needed — both paths were already correct and tested; added
`test/agent/agent.test.ts` regression coverage for the config-entry
case specifically.
- kilo (trivial): simplified the now single-element
`for (const profile of [BUILDER_PROFILE])` loop in
`test/altimate/prompt-profiles.test.ts` to a direct assertion on
`BUILDER_PROFILE`.
- cubic (P3): fixed the stale `session/termination.ts` doc comment that
still referred to "builder and builder-derived agents" — only builder
carries the completion contract now.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VqnuBDGkh1ZT65Ti7e6DHZ
Bugbot couldn't run - usage limit reachedBugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit. A user or team admin can review and increase usage limits in the Cursor dashboard. (requestId: serverGenReqId_68b5a820-f416-4498-9c7c-31ee1fc0d48a) |
|
👋 This PR was automatically closed by our quality checks. Common reasons:
If you believe this was a mistake, please open an issue explaining your intended contribution and a maintainer will help you. |
3 similar comments
|
👋 This PR was automatically closed by our quality checks. Common reasons:
If you believe this was a mistake, please open an issue explaining your intended contribution and a maintainer will help you. |
|
👋 This PR was automatically closed by our quality checks. Common reasons:
If you believe this was a mistake, please open an issue explaining your intended contribution and a maintainer will help you. |
|
👋 This PR was automatically closed by our quality checks. Common reasons:
If you believe this was a mistake, please open an issue explaining your intended contribution and a maintainer will help you. |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: f61cd61a16
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".
…of throwing Second facet of the P1 review thread on the data-qa removal (codex `agent.ts:320`, second thread — "Migrate the removed data-qa default to analyst"). #1217 let `default_agent: "data-qa"` alone — with no matching `agent.data-qa` config entry — opt into the native data-qa profile. Removing that registration left `defaultInfo()` throwing `default agent "data-qa" not found` for anyone whose persisted config still sets `default_agent: "data-qa"`. Every call site that resolves the default agent when none is supplied (`session/prompt.ts`, the session HTTP routes, ACP) depends on `defaultAgent()`/`defaultInfo()` succeeding, so an upgrading user in this state could no longer start an ordinary default-agent session — a real strand, not a cosmetic one. Fix: in `defaultInfo()`, when `default_agent === "data-qa"` and no agent actually resolves for that name, fall back to `analyst` — the documented replacement for read-only data questions — with a one-time warning logged instead of a hard failure. A user who separately defines their own `agent.data-qa` config entry is unaffected: that legitimate custom agent resolves first, before the fallback is ever considered (see the companion test). This is scoped to the literal removed name "data-qa" — it does not change the existing, correct behavior for a generic invalid/typo'd `default_agent` (still throws; see the pre-existing "defaultAgent throws when default_agent points to non-existent agent" test, unchanged). Added two tests to `test/agent/agent.test.ts`: - `default_agent: "data-qa"` with no matching agent entry resolves to `analyst` (not a throw). - `default_agent: "data-qa"` WITH an explicit `agent.data-qa` config entry still resolves to that user-defined agent, not the migration. `bun run typecheck` clean; `bun test test/agent/` 53/53 pass (plus the 2 new tests); builder byte-identity pin untouched. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VqnuBDGkh1ZT65Ti7e6DHZ
The marker guard (bun run script/upstream/analyze.ts --markers) flagged `let agent = agents[c.default_agent]` in agent.ts's defaultInfo() as unmarked custom code — the previous commit's altimate_change block started one line too late, after the const->let conversion needed for the data-qa default-agent migration. Moved the marker start to cover that line too. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VqnuBDGkh1ZT65Ti7e6DHZ
Bugbot couldn't run - usage limit reachedBugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit. A user or team admin can review and increase usage limits in the Cursor dashboard. (requestId: serverGenReqId_f435ad1f-a6e0-4ac9-ab29-929bd93e0af8) |
|
👋 This PR was automatically closed by our quality checks. Common reasons:
If you believe this was a mistake, please open an issue explaining your intended contribution and a maintainer will help you. |
2 similar comments
|
👋 This PR was automatically closed by our quality checks. Common reasons:
If you believe this was a mistake, please open an issue explaining your intended contribution and a maintainer will help you. |
|
👋 This PR was automatically closed by our quality checks. Common reasons:
If you believe this was a mistake, please open an issue explaining your intended contribution and a maintainer will help you. |
|
👋 This PR was automatically closed by our quality checks. Common reasons:
If you believe this was a mistake, please open an issue explaining your intended contribution and a maintainer will help you. |
1 similar comment
|
👋 This PR was automatically closed by our quality checks. Common reasons:
If you believe this was a mistake, please open an issue explaining your intended contribution and a maintainer will help you. |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: b6392fdb3d
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".
| log.warn( | ||
| 'the "data-qa" agent was removed; defaulting to "analyst" for read-only data questions — set default_agent to override', | ||
| ) |
There was a problem hiding this comment.
When a user upgrades with default_agent: "data-qa", this warning is sent through the fork-local Log shim, whose output is disabled unless ALTIMATE_PRINT_LOGS or OPENCODE_PRINT_LOGS is explicitly enabled (src/altimate/util/log.ts). Normal CLI and TUI users therefore receive no warning that their configured writable agent was silently replaced by the read-only analyst, and the configuration remains stale, so the silent substitution repeats on later starts. Emit this through a user-visible diagnostic channel or persistently migrate the setting.
Useful? React with 👍 / 👎.
There was a problem hiding this comment.
1 issue found across 2 files (changes from recent commits).
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="packages/opencode/src/agent/agent.ts">
<violation number="1" location="packages/opencode/src/agent/agent.ts:618">
P3: The advertised one-time migration warning never reaches default TUI users. `log.warn` goes through the `Log` shim (`src/altimate/util/log.ts`), whose `shouldLog` requires `printEnabled()` — it only writes when `OPENCODE_PRINT_LOGS`/`ALTIMATE_PRINT_LOGS` is set, and that is deliberately OFF for the in-process TUI server. So an upgrading user whose persisted `default_agent: "data-qa"` is silently redirected to `analyst` will not see the notice that the PR description promises ("Fall back to `analyst` ... with a one-time warning"). Consider surfacing the migration through a channel that reaches interactive users (e.g. a session/notification) rather than the print-gated stderr log.</violation>
</file>
Tip: Review your code locally with the cubic CLI to iterate faster.
Re-trigger cubic
| if (!warnedRemovedDataQaDefault) { | ||
| warnedRemovedDataQaDefault = true | ||
| log.warn( | ||
| 'the "data-qa" agent was removed; defaulting to "analyst" for read-only data questions — set default_agent to override', |
There was a problem hiding this comment.
P3: The advertised one-time migration warning never reaches default TUI users. log.warn goes through the Log shim (src/altimate/util/log.ts), whose shouldLog requires printEnabled() — it only writes when OPENCODE_PRINT_LOGS/ALTIMATE_PRINT_LOGS is set, and that is deliberately OFF for the in-process TUI server. So an upgrading user whose persisted default_agent: "data-qa" is silently redirected to analyst will not see the notice that the PR description promises ("Fall back to analyst ... with a one-time warning"). Consider surfacing the migration through a channel that reaches interactive users (e.g. a session/notification) rather than the print-gated stderr log.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At packages/opencode/src/agent/agent.ts, line 618:
<comment>The advertised one-time migration warning never reaches default TUI users. `log.warn` goes through the `Log` shim (`src/altimate/util/log.ts`), whose `shouldLog` requires `printEnabled()` — it only writes when `OPENCODE_PRINT_LOGS`/`ALTIMATE_PRINT_LOGS` is set, and that is deliberately OFF for the in-process TUI server. So an upgrading user whose persisted `default_agent: "data-qa"` is silently redirected to `analyst` will not see the notice that the PR description promises ("Fall back to `analyst` ... with a one-time warning"). Consider surfacing the migration through a channel that reaches interactive users (e.g. a session/notification) rather than the print-gated stderr log.</comment>
<file context>
@@ -592,7 +599,28 @@ export const layer = Layer.effect(
+ if (!warnedRemovedDataQaDefault) {
+ warnedRemovedDataQaDefault = true
+ log.warn(
+ 'the "data-qa" agent was removed; defaulting to "analyst" for read-only data questions — set default_agent to override',
+ )
+ }
</file context>
|
@anandgupta42 heads-up: (e.g. run https://github.com/AltimateAI/altimate-code/actions/runs/33850616707 — an unrelated PR failing only on these.) Since PR checks run on the merge-ref, this is currently blocking every open PR from going green. Could you take a look? |
2 participants
Issue for this PR
Closes #
Follow-up to #1217 (merged to
main), acting on the post-merge review decision to drop thedata-qaagent it introduced.Type of change
What does this PR do?
Removes the opt-in
data-qaagent added in #1217 and makes the existinganalystagent the documented data-question agent instead.Why drop
data-qa:analyst, but unsafely.analystis already described as "Read-only data exploration and analysis. Cannot modify files or run destructive SQL."data-qacovered the same use case (asking questions about data) but shipped with full read/writebuilderpermissions — a slim-prompt agent that can still run destructive SQL. That is the exact safety concern that got PR fix: scope the builder Pre-Execution Protocol to task shape #1215 closed. There's no reason to carry a second, less-safe path to the same jobanalystalready does safely.data-qawas an internal ablation showing that dropping the dbt-build packs is score-neutral on read-only data-QA workloads. I readanalyst.txt(packages/opencode/src/altimate/prompts/analyst.txt) to check whether it needed the same trim — it's already lean: no dbt build/finish-protocol content, no Pre-Execution Protocol pack, nothing the ablation targeted.analystalready embodies the insight. I leftanalyst.txtunchanged — the safe default the task called for when the prompt has no removable weight.What's kept: #1217's real contribution — splitting the monolithic
builder.txtinto an invariantcore.txt+ named packs underbuilder/packs/, assembled byte-identically byprofiles.ts— is untouched.BUILDER_PROFILE,PROMPT_BUILDER,assemble, andFRAGMENTSare unmodified, and the builder byte-identity pin (sha256 17663410dd9a…intest/altimate/prompt-identity.ts) stays green.Changes:
packages/opencode/src/agent/agent.ts— removed the conditionaldata-qaregistration block (theALTIMATE_DATA_QA_PROFILEenv flag,cfg.agent["data-qa"]config-entry opt-in, andcfg.default_agent === "data-qa"gate), and the now-unusedFlagimport.packages/opencode/src/altimate/prompts/profiles.ts— removedDATA_QA_PROFILEandPROMPT_DATA_QA.BUILDER_PROFILE/PROMPT_BUILDER/assemble/FRAGMENTSare untouched.packages/opencode/src/session/termination.ts— removeddata-qafromCOMPLETION_CONTRACT_AGENTS(now{"builder"}only).packages/opencode/test/agent/data-qa-profile.test.ts.packages/opencode/test/altimate/prompt-profiles.test.ts— removed the "data-qa profile composition"describeblock and theDATA_QA_PROFILE/PROMPT_DATA_QAimports; kept the builder byte-identity and cross-process determinism tests.packages/opencode/test/session/termination.test.ts— removed thedata-qaassertions from the completion-contract test.packages/opencode/test/altimate/prompt-identity.ts— updated a stale doc comment referencing the deleted test file.data-qaorALTIMATE_DATA_QA_PROFILE(that landed only in code in refactor: split builder prompt into invariant core + named packs (byte-identical assembly) + opt-in data-qa profile #1217), so there was nothing to remove. Updateddocs/docs/configure/agents.md,docs/docs/data-engineering/agent-modes.md,docs/docs/getting-started.md, anddocs/docs/getting-started/quickstart.mdto explicitly describeanalystas the agent for asking questions about your data (in addition to its existing "read-only exploration" framing), so the safe choice for "just ask questions about my data" is unambiguous.builder(full read/write) andplan(planning-only) descriptions were left as-is; they were already accurate.A repo-wide grep for
data-qa,DATA_QA,ALTIMATE_DATA_QA_PROFILE, andPROMPT_DATA_QAafter these changes returns no results.Review-round follow-up (upgrade-safety on the two legacy-config paths): codex and cubic flagged what happens to a config that still references the removed agent after an upgrade. Traced both paths and handled each on its actual behavior:
agent: {"data-qa": {...}}config entry — traced the config→agent registration path inagent.ts: it falls through to the existing generic custom-agent path (already covered by"custom agent from config creates new agent"), producing a fully-formed, non-crashing agent (mode: "all",native: false). No migration code needed — verified with a new regression test ("a leftover config agent.data-qa entry resolves as a harmless generic custom agent, not a crash") rather than adding a special case for a key that was never in a release.default_agent: "data-qa"with no matchingagent.data-qaentry — this one did need a fix. refactor: split builder prompt into invariant core + named packs (byte-identical assembly) + opt-in data-qa profile #1217 explicitly special-cased this exact configuration to register the native profile, sodefaultInfo()would otherwise throwdefault agent "data-qa" not foundand strand every call site that resolves the default agent (session/prompt.ts, session HTTP routes, ACP). Fixed by falling back toanalyst— the documented replacement — with a one-time warning, scoped to the literal name"data-qa"(a generic invalid/typo'ddefault_agentstill throws, unchanged). Two new tests cover the fallback and the case where a user's ownagent.data-qaentry should win over the migration.How did you verify your code works?
bun run typecheck(repo root, via turbo) — clean.bun test test/agent/ test/altimate/ test/session/(packages/opencode) — 6119 pass, 0 fail from this change. Twotest/session/prompt.test.tstests (loop calls LLM and returns assistant message,loop surfaces content-filter finishes as session errors) are flaky/timeout-prone against the mocked LLM harness and reference agent"build"(the pre-existingbuilderalias) — nothing to do withdata-qa; the adjacent test in that file already carries a standingBUG: REGRESSIONcomment about this exact flakiness. Reproduces identically before and after this change.bun test test/altimate/prompt-profiles.test.tsspecifically — 4/4 pass, including the builder byte-identity assertion against the pinnedsha256 17663410dd9a….bun run script/upstream/analyze.ts --markers --base origin/main --strict— clean (All custom code in upstream-shared files is properly marked); removing thedata-qablock shrank the markedaltimate_changeregion inagent.ts, it did not need a new one.data-qa/DATA_QA/ALTIMATE_DATA_QA_PROFILE/PROMPT_DATA_QA— zero remaining references.f61cd61a16,e3e47bc6ca,b6392fdb3d) address the review round above:bun test test/agent/ test/altimate/prompt-profiles.test.ts test/altimate/prompt-identity.ts test/session/termination.test.ts— 92/93 pass, 1 pre-existing todo, 0 fail (includes the 3 new regression tests for the legacy-config paths);bun run typecheckclean;bun run script/upstream/analyze.ts --markers --base origin/main --strictclean.Screenshots / recordings
N/A — no UI change.
Checklist
Note
Medium Risk
Changes default-agent resolution for users who still have
default_agent: "data-qa"in config; migration to analyst is intentional and tested, but it alters which permissions and prompt users get on upgrade.Overview
Removes the opt-in
data-qanative agent (env flag, config registration, and slimPROMPT_DATA_QAprofile) and positionsanalystas the supported read-only path for asking questions about data.Runtime behavior:
default_agent: "data-qa"with no matchingagent.data-qablock now resolves toanalystwith a one-time log warning instead of failing startup; a leftoveragent.data-qaconfig entry becomes a generic custom agent. Run-modeDONEcompletion instructions apply only tobuilder(data-qadropped from the completion contract).Docs: Agent and getting-started pages now describe Analyst explicitly for data Q&A. Builder prompt assembly and byte-identity pins are unchanged.
Tests: Deletes
data-qa-profiletests and adds regressions for the migration paths above.Reviewed by Cursor Bugbot for commit b6392fd. Bugbot is set up for automated code reviews on this repo. Configure here.
Summary by cubic
Removes the opt-in
data-qaagent and makes the existinganalystagent the documented agent for asking questions about your data.data-qaduplicatedanalyst's purpose while shipping with full read/write builder permissions, and its insight was already embodied inanalyst's lean prompt. The builder prompt split into core + named packs is untouched.Changes
data-qaregistration, its env-flag/config opt-ins,DATA_QA_PROFILE, and the completion-contract entry.default_agent: "data-qa"now resolves toanalystwith a one-time warning instead of stranding default-agent sessions; an explicitagent.data-qaconfig entry still wins.agent.data-qaconfig entry resolves as a harmless generic custom agent.analystas the agent for asking questions about your data.Written for commit b6392fd. Summary will update on new commits.
Summary by CodeRabbit
Documentation
Changes