// Sync the findings record against local review artifacts and optionally a GitHub PR. Accept optional `path` and GitHub PR `url`; when `url` is provided, sync against both remote PR state and local state, otherwise default to local-only sync. Default to `path=infer`. Confirm effective variables before starting.
| name | sync-findings |
| description | Sync the findings record against local review artifacts and optionally a GitHub PR. Accept optional `path` and GitHub PR `url`; when `url` is provided, sync against both remote PR state and local state, otherwise default to local-only sync. Default to `path=infer`. Confirm effective variables before starting. |
Read these shared references when needed:
../shared/references/findings.schema.md../shared/references/findings.lifecycle.mdKeep the active findings record aligned with current local state and, when requested, current GitHub PR state.
Accept an optional GitHub PR url from the prompt.
Default:
url=local-onlyAccept an optional path from the prompt.
path is provided, use it as the local design or findings folder.path is omitted, infer it from the branch, subsystem, or matching docs and state the inferred value before starting.Default:
path=inferEstablish sync scope. Confirm the effective variables first:
pathurlLoad the active findings record.
Use path/findings.md.
SAVE FIRST — write every new finding into path/findings.md before doing anything else.
This is the non-negotiable first action of every sync. Before proposing fixes, discussing implications, replying to threads, or updating any other doc:
## Open Findings using the canonical schema (ID, severity, files, evidence, decision pending).needs-user-decision unless the user has already given direction.## Summary section so the latest state is visible at the top.findings.md is up to date may you start summarizing for the user or proposing fixes.This rule exists because diving straight into "let me explain / propose / fix" mode loses context and leaves the user to reconstruct the finding list themselves. The doc is the source of truth, not the chat history.
Sync sources into findings.
Re-check existing findings against current local code and docs.
If url is present, pull open PR comments and thread state into the same record.
Ask clarifying questions before parking ambiguity. If a finding's disposition or intended action is still unclear, ask the next concrete question instead of silently leaving it vague.
Close only what is clearly closed. When a GitHub thread maps to a closed finding, leave a short reply with the finding ID when relevant and resolve the thread. Leave still-open or still-ambiguous threads untouched.
triage-findings when the user needs help deciding what to do next.Audit and update the supported LLM model list in assets.py against litellm's registry (models.litellm.ai). Use when adding new models, pruning outdated ones, or verifying the list is correct.
Update the API reference documentation by downloading the latest OpenAPI spec from production and regenerating the Docusaurus API docs
Resolve findings by implementing the chosen fix path in code, tests, or docs. Accept optional `path` and a `priority` selector; by default resolve only the next highest remaining priority bucket, in order `P0`, `P1`, `P2`, `P3`. Also accept explicit levels or `all`. Default to `path=infer`. Confirm effective variables before starting.
Perform a fresh-context scan of code and docs that turns verification observations and missing-test gaps into findings. Accept optional `path` and `depth` parameters and default to `path=infer`, `depth=deep`. Confirm effective variables before starting.
Run or inspect the relevant validation paths and turn failures, regressions, or missing coverage into findings. Accept optional `path` and `depth` parameters and default to `path=infer`, `depth=deep`. Confirm effective variables before starting.
Coordinate findings work with the user, decide whether scan, test, or sync should run, and turn the current findings set into a ready plan. Accept optional `path` and GitHub PR `url`; default to `path=infer`. Confirm effective variables before starting.