docs(spx-v2): correct a startup claim STATE.md asserted without running it - #15
Merged
Merged
Conversation
…ing it STATE.md told every reader that a leftover data_source.vix.allow_cboe_vix3m_fallback makes the process refuse to start. Measured on the Mac in PR PrimeIntellect-ai#291: it does not. extra="forbid" fires inside provider composition and becomes `startup: GONE data (paper_data_composition_refused)`; with a live Gateway the process runs and binds. Raw-YAML startup refusals come from settings_adapter.py's STRATEGY_OWNED_PATHS, which this key is not on. The claim came from PR PrimeIntellect-ai#270's body and the production_data.py:89 comment, both of which assert it, and was copied into this file, the Mac dispatch prompt and several summaries without being executed once. That is exactly LESSONS.md 13 -- never publish a claim you have not run in the form you state it -- broken by the file that carries the rule. The code comment remains wrong and deserves its own fix lane. Also records the first runtime-proof attempt honestly: PR PrimeIntellect-ai#291 did not pass. Zero clean passes, two measured FAILs, three observables unprovable without a live process, because IB Gateway was down and the PAPER runtime fail-closes at adapter start. And adds the quoting rule: execution.mode must be "off" quoted, since bare off is YAML False. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01G5B7QM1QLQuWSBMCxiCzS6
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. |
Owner
Author
|
![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: e51d092a3f
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".
…ctions Codex found four P2s on the correction PR. All verified, all valid. 1. The input of record still told the operator `execution.mode: off` unquoted -- the exact form this PR establishes the loader refuses. An implementing agent following it would reproduce the refusal. Now quoted, with the reason. 2. I over-corrected. Having found that a leftover allow_cboe_vix3m_fallback does NOT block startup, I demoted deleting it to "housekeeping". Wrong in the other direction: it still leaves `startup: GONE data (paper_data_composition_refused)` with entries IMPOSSIBLE, so a healthy /readiness is unreachable while it is present. Only the REASON was wrong; the step is still required. 3. The operator steps implied that starting the Gateway yields healthy session domains. It does not. Observable (c) failed artifact_missing on BOTH domains, and those artifacts come from the Hermes publisher, not from Gateway uptime. The steps now name three preconditions, not one. 4. Bare `PR PrimeIntellect-ai#291` in prime-agent markdown autolinks to prime-agent's own PrimeIntellect-ai#291, which exists and is unrelated -- pointing readers away from the evidence the claims rest on. Now explicit cross-repository links. Findings 2 and 3 are the useful ones: a correction that swings past the truth is still wrong, and a checklist that names one precondition out of three promises an outcome it cannot deliver. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01G5B7QM1QLQuWSBMCxiCzS6
…oes not The operator had to say this twice tonight, which means the file that exists to stop repetition did not contain it. Now it is the first thing in LESSONS.md. Records what drift actually looked like in the session that did it, because the abstract rule was already known and did not prevent it. It never announced itself; it looked like being helpful -- the operator said the repo was unclean, so files got edited instead of a brief getting filed (spx#290); a stale doc blocked the operator, so the doc got fixed directly instead of a brief getting filed (spx#292). Neither touched a .py file, which is exactly why it felt allowed. The order says all code, any repository, and product-repo files are the fleet's lane whatever their extension. The tell is written down as a sentence to catch in the act: "this is small, I can just do it." Small is what the fleet is for. Also records the one legitimate exception and its exact shape -- an operator prompt that re-designates the session as Prime Agent IS a dispatch, it covers the briefs named in that prompt and nothing else, and the session reverts to observer when they are delivered rather than carrying the designation forward. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01G5B7QM1QLQuWSBMCxiCzS6
4 tasks
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
2 participants
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.
What
Corrects a factual error in
orchestration/spx-v2/STATE.md, and records the first runtime-proof attempt honestly.The error
STATE.mdtold every reader — and the Mac dispatch prompt repeated it — that a leftoverdata_source.vix.allow_cboe_vix3m_fallbackmakes the process refuse to start.Measured on the Mac in spx-0dte-bot-v2 PR PrimeIntellect-ai#291: it does not.
extra="forbid"fires inside provider composition and is converted tostartup: GONE data (paper_data_composition_refused)with "entries are IMPOSSIBLE" — with a live Gateway the process runs and binds in that state. Raw-YAML startup refusals come fromsettings_adapter.py's namedSTRATEGY_OWNED_PATHS, which this key is not on.Where it came from, since that matters more than the fix
PR PrimeIntellect-ai#270's body asserts the refusal, and the
production_data.py:89comment asserts it too. I copied it from them intoSTATE.md, into the MacBook dispatch prompt, and into several summaries — without executing it once.That is
LESSONS.md§13 — never publish a claim you have not run in the form you state it — broken by the same author, in the same folder, four hours after writing the rule. The correction is recorded inSTATE.mdunder a new "Corrected claims" heading rather than silently patched, because the provenance is the useful part.The
production_data.py:89comment is still wrong and deserves its own small fix lane in the V2 repo. Deleting the key remains correct housekeeping; it is simply not a startup blocker.Also recorded
artifact_missingon both domains; no listener on 7497), three observables unprovable without a live process. IB Gateway was down, so the PAPER runtime fail-closes at adapter start and/readinesswas never presented. Bringing the Gateway up is operator-only (IB Key 2FA).execution.modemust be quoted — bareoffis YAMLFalseand the loader refuses it by design. Setexecution.mode: "off".Testing
Docs only. The corrected claim is sourced from a measurement in PR PrimeIntellect-ai#291's evidence file rather than from another document — which is the whole point of the change.
🤖 Generated with Claude Code
https://claude.ai/code/session_01G5B7QM1QLQuWSBMCxiCzS6
Generated by Claude Code