Skip to content

feat: add Pangram 4.1 short-text interpretation guidance - #63

Merged
Microck merged 4 commits into
mainfrom
feat/pangram-4-1-guidance
Oct 19, 2026
Merged

Microck merged 4 commits into
mainfrom
feat/pangram-4-1-guidance

Conversation

@Microck

@Microck Microck commented Oct 19, 2026

Copy link
Copy Markdown
Owner

Summary

  • Investigate the complete official 4.1 announcement, including the follow-up posts beyond the supplied screenshot: calibration/ranges, under-150-word caution, and reported false-positive improvements.
  • Add shared short-text guidance to Pretty/Markdown and TUI results, without changing machine-readable schemas or detector scores.
  • Record verified sources and the remaining public API contract gap. No invented selector, numerical interval, or relaxed response-version validation.

Scope and dependency

Stacked on #62 (qa/documentation-browser-flows) to reuse its dependency/security and browser-test fixes. This PR is not a claim of complete 4.1 wire compatibility: public docs and SDK still show the 4.0 contract and do not document the new dashboard range fields. Server-side accuracy claims are not independently benchmarked here.

Verification

  • Full Rust all-features suite: 979 passed, 3 ignored, 0 failed
  • Real compiled TUI / loopback HTTP / PTY journeys: 13 passed; short-text notice explicitly asserted and its rendered screen inspected
  • Documentation generation, link checks, typecheck and production build passed
  • Formatting and diff checks passed; human-output goldens and TUI snapshot updated
  • Threshold cases: 149/150 words, multi-segment total, absent segments, overflow
  • No paid API calls, provider credits, dashboard scraping, public links, or releases

Sources and precise limitations: docs/pangram-4-1-investigation.md.

@vercel

vercel Bot commented Oct 19, 2026 •

Copy link
Copy Markdown

The latest updates on your projects. Learn more about Vercel for GitHub.

Project Deployment Actions Updated
pangram-cli Ready Ready Preview Oct 19, 2026 10:27pm UTC

@coderabbitai

coderabbitai Bot commented Oct 19, 2026 •

Copy link
Copy Markdown

Review in Change Stack →

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info
⚙️ Run configuration
  • Configuration used: Organization UI
  • Review profile: ASSERTIVE
  • Plan: Advanced
  • Run ID: b8c707a0-7daa-41bf-9700-50bc76f2e6b4
📥 Commits

Reviewing files that changed from the base of the PR and between 4d4a0ec and 6032aae.

⛔ Files ignored due to path filters: 1
  • tools/tui-acceptance/__snapshots__/tui-acceptance.test.ts.snap is excluded by !**/*.snap
📒 Files selected for processing: 18
  • .github/workflows/browser.yml
  • .tegami/pangram-4-1-guidance.md
  • docs-app/content/docs/index.mdx
  • docs-app/public/llms-full.txt
  • docs-app/public/markdown/index.md
  • docs-app/tsconfig.e2e.json
  • docs-app/tsconfig.json
  • docs/contracts.md
  • docs/pangram-4-1-investigation.md
  • src/domain.rs
  • src/domain/detection_guidance.rs
  • src/output/projection.rs
  • src/tui/result-lines.rs
  • tests/projection-contract.rs
  • tests/support/golden/markdown_success.md
  • tests/support/golden/pretty_success.txt
  • tests/support/golden/pretty_success_color.txt
  • tools/tui-acceptance/tui-acceptance.test.ts

Included review availability: This review used your included allowance. Your plan provides up to 1 included review per hour; 0 remain after this review.

This PR adds a caution to human-readable AI-detection results when validated segment metadata reports fewer than 150 words; it does not change scores or machine-readable output.

  • The caution requires a nonempty segment list, matching segment counters, and a checked word-count total below 150; missing, inconsistent, or overflowing metadata does not trigger it.
  • The CLI does not infer undocumented Pangram 4.1 API fields or numerical ranges.
  • Tests cover the threshold and output contract; test execution results were not supplied.

Walkthrough

The CLI now shows a short-text caution in Pretty, Markdown, and TUI AI results when the returned segments contain fewer than 150 words in total and the segment counters match. The caution does not change scores or machine-readable output. Documentation describes the limits of the Pangram 4.1 announcements and states that the CLI does not derive a dashboard percentage range from segment confidence.

Sequence Diagram(s)

sequenceDiagram
  participant PrettyMarkdown as Pretty and Markdown output
  participant Guidance as short_text_caution
  participant TUI as TUI result rendering
  PrettyMarkdown->>Guidance: Pass AI detection result
  Guidance-->>PrettyMarkdown: Return caution or None
  TUI->>Guidance: Pass AI detection result
  Guidance-->>TUI: Return caution or None
Loading

Priority: ⬇️ Low

Merge Risk: ⚪ Minimal · up to 6032a

The short-text guidance is covered across the human-readable views, and no issue remains that needs resolution before merging.

- [ ] Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

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.

❤️ Share

Comment @coderabbitai help to get the list of available commands.

Microck commented Oct 19, 2026

Copy link
Copy Markdown
Owner Author

@coderabbitai full review

@codex review

Please review the short-text threshold/aggregation, output compatibility, real PTY coverage and documented 4.1 API boundary. This is stacked on #62; implementation is limited to verified public announcement guidance, not undocumented wire fields.

@coderabbitai

coderabbitai Bot commented Oct 19, 2026 •

Copy link
Copy Markdown
✅ Action performed

Full review finished.

@chatgpt-codex-connector

chatgpt-codex-connector Bot commented Oct 19, 2026 •

Copy link
Copy Markdown

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review Status Commit Review trigger
📝 Code Review ✅ Completed 2026-10-19T22:17:42.147364Z 6032aae Manual request
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" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

@chatgpt-codex-connector

Copy link
Copy Markdown

Codex Review: Didn't find any major issues. Can't wait for the next one!

Reviewed commit: c391936eeb

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".

Microck commented Oct 19, 2026

Copy link
Copy Markdown
Owner Author

Vercel preview exposed a production-build boundary issue: docs-app's tsconfig included browser test files whose packages are only root tooling dependencies. Production now excludes those files, while a separate tsconfig.e2e.json is explicitly typechecked in browser CI. Local isolated typecheck and docs production build pass.

@coderabbitai full review
@codex review

@coderabbitai

coderabbitai Bot commented Oct 19, 2026 •

Copy link
Copy Markdown
⚠️ Action not completed

Review rate limited.


Your included review limit is currently reached under our Fair Usage Limits Policy. This review may still proceed through usage-based billing if eligible. Your next included review will be available in 59 minutes.

@chatgpt-codex-connector

Copy link
Copy Markdown

Codex Review: Didn't find any major issues. Already looking forward to the next diff.

Reviewed commit: dbd7dc69e2

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".

Microck commented Oct 19, 2026

Copy link
Copy Markdown
Owner Author

@coderabbitai full review

Retrying the current-head review after the quoted included-review cooldown. Please report any remaining actionable findings.

@coderabbitai

coderabbitai Bot commented Oct 19, 2026 •

Copy link
Copy Markdown
⚠️ Action not completed

Review rate limited.


Your included review limit is currently reached under our Fair Usage Limits Policy. This review may still proceed through usage-based billing if eligible. Your next included review will be available in 59 minutes.

@Microck
Microck changed the base branch from qa/documentation-browser-flows to main October 19, 2026 22:04

Microck commented Oct 19, 2026

Copy link
Copy Markdown
Owner Author

The prerequisite #62 is now merged, with all 28 returned checks and Vercel preview passing and both current-head reviews clean. Retargeted this PR to main. Please review the remaining Pangram 4.1 guidance delta against main.

@coderabbitai full review
@codex review

@coderabbitai

coderabbitai Bot commented Oct 19, 2026 •

Copy link
Copy Markdown
⚠️ Action not completed

Review rate limited.


Your included review limit is currently reached under our Fair Usage Limits Policy. This review may still proceed through usage-based billing if eligible. Your next included review will be available in 57 minutes.

@chatgpt-codex-connector chatgpt-codex-connector Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: dbd7dc69e2

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".

Comment thread src/domain/detection_guidance.rs Outdated

Microck commented Oct 19, 2026

Copy link
Copy Markdown
Owner Author

@coderabbitai full review
@codex review

Latest fix validates complete segment metadata before making a short-document claim; documented contract and regression tests updated.

@coderabbitai

coderabbitai Bot commented Oct 19, 2026 •

Copy link
Copy Markdown
✅ Action performed

Full review finished.

@chatgpt-codex-connector

Copy link
Copy Markdown

Codex Review: Didn't find any major issues. Nice work!

Reviewed commit: 6032aae232

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".

@Microck
Microck merged commit c4d8bed into main Oct 19, 2026
29 checks passed

This branch was successfully deployed

1 active deployment
Preview — 6032aae2 Deployed Oct 19, 2026 by vercel[bot]
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant