Skip to content

feat(scoring): simulate open-pr pressure scenarios - #348

Merged
JSONbored merged 2 commits into
JSONbored:mainfrom
Lobster-0429:feat/open-pr-pressure-scenarios
Jun 3, 2026
Merged

feat(scoring): simulate open-pr pressure scenarios#348
JSONbored merged 2 commits into
JSONbored:mainfrom
Lobster-0429:feat/open-pr-pressure-scenarios

Conversation

@Lobster-0429

@Lobster-0429 Lobster-0429 commented Jun 3, 2026

Copy link
Copy Markdown
Contributor

Summary

Scope

  • This PR is focused and does not mix unrelated backend, UI, MCP, docs, dependency, and deploy changes.
  • This follows CONTRIBUTING.md and does not reintroduce GitHub Pages, VitePress, site/, or CNAME.
  • I linked an issue, or this is small enough that the summary explains why an issue is not needed.

Validation

  • git diff --check
  • npm run actionlint
  • npm run typecheck — clean
  • npm run test:coverage locally — 788 pass (1 skipped); pre-existing Windows failures confirmed on main before this branch; coverage stays above 97%
  • npm run test:workers
  • npm run build:mcp
  • npm run test:mcp-pack
  • npm run ui:openapi:check
  • npm run ui:lint
  • npm run ui:typecheck
  • npm run ui:build
  • npm audit --audit-level=moderate
  • New or changed behavior has unit/integration tests for new branches, fallback paths, and sanitizer boundaries

If any required check was skipped, explain why:

  • UI/MCP/worker checks not applicable — this PR adds one pure simulation service plus its unit test; no UI, MCP, or worker surface changed

Safety

  • No secrets, wallet details, hotkeys, coldkeys, user PATs, private keys, raw trust scores, private rankings, or private maintainer evidence are exposed.
  • Public GitHub text stays sanitized, low-noise, and does not imply compensation guarantees or optimization tactics — a test asserts no payout/reward/score claims appear.
  • Auth, cookie, CORS, GitHub App, Cloudflare, or session changes include negative-path tests. (not applicable)
  • API/OpenAPI/MCP behavior is updated and tested where needed. (simulator is pure; wiring into surfaces is out of scope)
  • Visible UI changes include screenshots or a short recording. (not applicable)
  • Public docs/changelogs are updated where needed. (not applicable)

Notes

OpenPrPressureSimulation shape:

Field Meaning
lane contributor or maintainer (from RoleContext.maintainerLane)
queuePressure low / medium / high / critical, or unknown when queueHealth is null
recommendedOption the rank-1 option
scenarios[] each option with rank, recommended, facts, assumptions, tradeoffs, blockers
summary one-sentence public-safe recommendation

Contributor-lane ranking:

Own open PRs Pressure Order (rank 1 → 3)
> 0 high/critical/unknown cleanup_first, wait, open_new_work
> 0 low/medium cleanup_first, open_new_work, wait
0 high/critical/unknown wait, open_new_work, cleanup_first
0 low/medium open_new_work, wait, cleanup_first

cleanup_first carries a blocker ("no open PR(s)… nothing to clean up") when the contributor has no open PRs.

Maintainer-lane ranking: open_new_work by default (maintainer work is repo-health work, not outside-contributor queue load), switching to cleanup_first (queue triage) under critical pressure. Maintainers are never penalized for their own concurrent PRs (no blocker emitted).

Facts vs assumptions: facts are observed queue signals (open PR/issue counts, stale PRs, the contributor's own open PR count); assumptions are projections (e.g. "opening another PR would add to the current high pressure"). Missing-signal repos mark these as estimates and recommend a conservative wait.

Test structure (16 tests): four repo fixtures (low/high/maintainer/missing-signal) asserting recommended option, full ranking order, facts, assumptions, and blockers; a facts-vs-assumptions separation check; and two sanitizer tests confirming every field passes sanitizePublicComment and no payout/reward/score language appears.

@Lobster-0429
feat(scoring): simulate open-pr pressure scenarios
c2ae95d 
Adds src/services/open-pr-pressure-scenarios.ts, a pure simulator that
models how opening another PR affects repo queue pressure and contributor
strategy. Compares three options — open new work, wait, or clean up
existing work first — using repo queue health and maintainer-lane signals.
Advisory only; no payout/reward/score claims and no GitHub actions.

simulateOpenPrPressure({ queueHealth, roleContext, contributorOpenPrCount })
returns a ranked OpenPrPressureSimulation:
- queuePressure: low/medium/high/critical, or "unknown" when queueHealth
  is null (missing-signal repos).
- lane: "maintainer" when roleContext.maintainerLane, else "contributor",
  ranked by separate heuristics.
- scenarios: each option carries rank, recommended flag, known facts
  (observed queue counts/state), assumptions (projections), tradeoffs,
  and blockers — facts and assumptions are kept in distinct fields.
- recommendedOption + summary.

Contributor ranking: own open PRs + heavy pressure -> cleanup_first;
own PRs + light pressure -> cleanup_first; no own PRs + heavy pressure
-> wait; otherwise open_new_work. cleanup_first carries a blocker when
the contributor has no open PRs to clean. Maintainer ranking: open_new_work
by default, cleanup_first (queue triage) under critical pressure; maintainers
are never penalized for their own concurrent PRs.

Scope is open-PR pressure only; linked-issue eligibility and duplicate/
stale blockers are handled by separate services (JSONbored#288/JSONbored#289). All public
text runs through sanitizePublicComment.

test/unit/open-pr-pressure-scenarios.test.ts (new, 16 tests):
- Fixtures for low-pressure, high-pressure, maintainer-lane, and
  missing-signal repos, asserting recommended option, ranking order,
  facts, assumptions, and blockers.
- Facts-vs-assumptions separation check.
- Public sanitizer tests: every scenario field across all fixtures passes
  sanitizePublicComment; no payout/reward/score claims appear.
@Lobster-0429
Lobster-0429 requested a review from JSONbored as a code owner June 3, 2026 09:45
@dosubot dosubot Bot added the size:L label Jun 3, 2026
JSONbored
JSONbored approved these changes Jun 3, 2026
View reviewed changes

@JSONbored JSONbored left a comment

Copy link
Copy Markdown
Owner

Choose a reason for hiding this comment

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

@web-dev0521 this is a focused addition and the shape looks right.

A few notes:

  • Keeping the scenario math isolated makes this easy to validate before wiring it into higher-level planning surfaces.
  • The tests cover the pressure cases directly instead of only checking a happy path.
  • The PR description has the issue and validation evidence needed for review.

No code changes requested.

Validation expected:

  • Keep the current CI suite green through merge.

@dosubot dosubot Bot added the lgtm label Jun 3, 2026
@JSONbored
JSONbored merged commit fab16a8 into JSONbored:main Jun 3, 2026
4 checks passed
@github-project-automation github-project-automation Bot moved this from Todo to Done in gittensory - v1 roadmap Jun 3, 2026
@JSONbored JSONbored added gittensor:feature Gittensor-scored feature linked to a feature issue — scores a 0.25x multiplier. and removed feature labels Jun 10, 2026
@oktofeesh1 oktofeesh1 mentioned this pull request Jun 19, 2026
Merged
oktofeesh1 added a commit to oktofeesh1/loopover-oktofeesh1 that referenced this pull request Jun 19, 2026
@oktofeesh1 @claude
fix(signals): wire open-PR pressure scenarios into the branch scenari…
77aceb4 
…o summary (JSONbored#348)

`simulateOpenPrPressure` (open-pr-pressure-scenarios.ts, JSONbored#348) and the
`renderPublicScenarioSummary` `options`/headline rendering for it were both
implemented and tested, but the only production caller —
buildLocalBranchAnalysis — never built or passed `pressureSimulation`. As a
result `scenarioSummary.options` was ALWAYS empty in analyze-branch / preflight
/ local-diff, and the strategy guidance (open new work / wait / clean up first)
never reached contributors.

Build the simulation at the call site (roleContext is already in scope; derive
queueHealth via buildCollisionReport + buildQueueHealth, and the contributor's
own open-PR count) and pass it through. Pure/read-only and public-safe — the
renderer already sanitizes every line.

Test: local-branch.test.ts now asserts scenarioSummary.options is populated
(3 ranked options, exactly one recommended, non-empty label + nextStep).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
JSONbored added a commit that referenced this pull request Jun 20, 2026
@oktofeesh1 @claude @JSONbored
fix(signals): wire open-PR pressure scenarios into the branch scenari…
e9cfe0e 
…o summary (#348) (#923)

`simulateOpenPrPressure` (open-pr-pressure-scenarios.ts, #348) and the
`renderPublicScenarioSummary` `options`/headline rendering for it were both
implemented and tested, but the only production caller —
buildLocalBranchAnalysis — never built or passed `pressureSimulation`. As a
result `scenarioSummary.options` was ALWAYS empty in analyze-branch / preflight
/ local-diff, and the strategy guidance (open new work / wait / clean up first)
never reached contributors.

Build the simulation at the call site (roleContext is already in scope; derive
queueHealth via buildCollisionReport + buildQueueHealth, and the contributor's
own open-PR count) and pass it through. Pure/read-only and public-safe — the
renderer already sanitizes every line.

Test: local-branch.test.ts now asserts scenarioSummary.options is populated
(3 ranked options, exactly one recommended, non-empty label + nextStep).

Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-authored-by: ghost <49853598+JSONbored@users.noreply.github.com>
@github-actions github-actions Bot mentioned this pull request Jun 20, 2026
Closed
12 tasks
@jaso0n0818 jaso0n0818 mentioned this pull request Jun 24, 2026
Closed
5 tasks
This was referenced Jul 2, 2026
Open
Open
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@JSONbored JSONbored JSONbored approved these changes

Assignees

@Lobster-0429 Lobster-0429

Labels

gittensor:feature Gittensor-scored feature linked to a feature issue — scores a 0.25x multiplier.

Projects

No open projects
gittensory - v1 roadmap
Status: Done

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

feat(scoring): simulate open-pr pressure scenarios

2 participants

@Lobster-0429 @JSONbored