scip-query 0.10.12 → 0.12.0

package/skills/scip-health-improve/SKILL.md

@@ -1,155 +0,0 @@
----
-name: scip-health-improve
-description: Autonomously raise a scip-query repository's health score with verified, bounded cleanup. Use when the user asks to improve, raise, or maximize the health score; clean up as much as reasonable; continue after setup; pursue perfect code; or keep working with minimal interaction until no safe confirmed work remains.
----
-
-# SCIP Health Improve
-
-Use this skill for an autonomous health-improvement campaign. The target is not
-score chasing: fix confirmed issues that make the codebase harder to understand,
-verify, or change, then stop when the remaining score movement is unsafe,
-unproven, blocked, or not worth the risk.
-
-A confirmed issue is a raw scip-query signal that the agent has checked against
-current code, graph evidence, or repository checks and classified as real,
-intentional, false positive, or blocked.
-
-## Non-Negotiables
-
-1. If the environment supports durable goals and the user asked for autonomous
-   health improvement, create or continue a concrete goal: raise the scip-query
-   health score as high as reasonably possible by fixing confirmed actionable
-   findings while preserving behavior.
-2. Start from current evidence. Run setup first only if the repository has not
-   been bootstrapped; otherwise refresh the index and collect current signals.
-3. Before editing code, tell the user the current health score and the first
-   confirmed batch you will address.
-4. Do not treat raw health output as actionable until you inspect the relevant
-   code, graph, or verification evidence.
-5. Prioritize worst confirmed offenders first: high risk, high confidence, low
-   blast radius, strong expected score movement, and clear verification path.
-6. Keep working batch by batch until no safe confirmed improvement remains, the
-   score stabilizes, or a real external decision blocks progress.
-7. Do not run `scip-query setup-ci` in this workflow. CI setup is intentionally
-   separate until that path is mature.
-
-## Evidence Baseline
-
-Collect the baseline before ranking:
-
-```bash
-scip-query doctor
-scip-query status --capabilities
-scip-query status --capabilities
-# If freshness is stale, missing, or unknown:
-# scip-query reindex
-scip-query health --json --full
-scip-query diff-gate --json
-scip-query capability-matrix --json
-```
-
-Unavailable capabilities are not clean results. Record them as unproven signal
-classes and fix missing local prerequisites when that is in scope.
-
-## Priority Ladder
-
-Work from the safest, highest-confidence improvements toward broader design
-work:
-
-1. Compiler-verified cleanup-plan deletion batches.
-2. Incomplete migrations and recent duplicate echoes.
-3. Unused params, unused imports, dead symbols, and isolated symbols.
-4. Broken or stale docs, config validation issues, and missing co-change
-   partners.
-5. Thin wrappers, passthrough layers, stale abstractions, and speculative
-   generality.
-6. React or Vue duplicate components, hooks, composables, and large
-   component/view pressure.
-7. Directory architecture and maintainability repairs only when the evidence is
-   strong and the blast radius is bounded.
-
-## Confirmation Sweep
-
-Run the relevant commands for the current repository and scope:
-
-```bash
-scip-query cleanup-plan --verify --json
-scip-query recent-duplicates --json --full
-scip-query incomplete-migration --json --full
-scip-query unused-params --json --full
-scip-query stale-abstractions --json --full
-scip-query wrapper-candidates --json --full
-scip-query passthrough-candidates --json --full
-scip-query doc-drift --json --full
-scip-query co-change --json --full
-```
-
-For frontend projects, add the relevant React or Vue maintainability commands.
-
-Confirm candidates with targeted probes before editing:
-
-```bash
-scip-query code <symbol-or-file>
-scip-query refs <symbol>
-scip-query fan-in <symbol>
-scip-query fan-out <symbol>
-scip-query affected <symbol> --json
-scip-query change-surface <file> --json --full
-scip-query co-change <file> --json --full
-```
-
-## Autonomous Loop
-
-Before the first edit, report:
-
-```markdown
-Health score: N/100
-
-First confirmed batch:
-- <finding> - <evidence> - <planned fix> - <verification>
-
-Remaining notable signals:
-- <signal> - <status: unconfirmed, intentional, false positive, blocked>
-```
-
-Then repeat:
-
-1. Apply one verified deletion batch or one small targeted refactor.
-2. Run the project check that proves the touched behavior.
-3. Run:
-   ```bash
-   scip-query status --capabilities
-   # If freshness is stale, missing, or unknown:
-   # scip-query reindex
-   scip-query diff-gate --json
-   scip-query health --json --full
-   ```
-4. Fix new diff-gate findings immediately unless they are documented,
-   intentional, or blocked.
-5. Update `docs/scip-query/health-dossier.md` and
-   `docs/scip-query/health-dossier.json` when they exist.
-6. Pick the next highest-priority confirmed item.
-
-Use `scip-query cleanup-apply --verified --batch <n>` for verified deletion
-batches. Use `--all` only with explicit human approval. Use `--force-dirty`
-only after inspecting the touched files and confirming existing edits are
-unrelated to the cleanup batch.
-
-## Stop Criteria
-
-Stop the campaign when one of these is true:
-
-- the health score stabilizes after a verified batch;
-- only intentional, false-positive, blocked, or unconfirmed items remain;
-- the next improvement requires a product, API, compatibility, or ownership
-  decision from the user;
-- a missing toolchain or unavailable capability prevents trustworthy
-  verification;
-- further work would be broad architectural redesign rather than bounded health
-  improvement.
-
-## Closeout
-
-End with the starting and final health scores, batches applied, important files
-changed, verification commands run, remaining accepted or blocked items, and the
-next highest-value follow-up if one exists.

package/skills/scip-health-improve/agents/openai.yaml

@@ -1,4 +0,0 @@
-interface:
-  display_name: "SCIP Health Improve"
-  short_description: "Autonomous verified health cleanup"
-  default_prompt: "Use $scip-health-improve to raise this repo health score as high as reasonably possible with verified cleanup."

package/skills/scip-query-setup/SKILL.md

@@ -1,170 +0,0 @@
----
-name: scip-query-setup
-description: Set up or refresh scip-query in a repository so agents can use compiler-resolved code intelligence effectively. Use when adopting scip-query in a repo, running the robust `scip-query setup` bootstrap, fixing stale setup, adding AGENTS.md guidance, checking capability coverage, generating the health dossier, configuring .scipquery.json, or calibrating locality architectural boundaries.
----
-
-# scip-query Setup
-
-Use this skill when a repository needs to become a reliable scip-query workspace, not merely a repo where the binary happens to run.
-
-A scip-query workspace is a source repository with a current SCIP index, project config, and agent guidance that make code references compiler-resolved instead of guessed from text search. Its defining trait is that agents can answer "where is this defined, who uses it, and what changes with it?" from the index before editing.
-
-A setup pass is the bounded workflow that installs or verifies the tool, creates the project index, writes agent instructions, and records repo-specific calibration. Its defining trait is that it turns one-off local success into a repeatable operating surface for future agents.
-
-## Setup Workflow
-
-1. Confirm the project root.
-
-```bash
-pwd
-git rev-parse --show-toplevel
-```
-
-2. Verify the command is available.
-
-```bash
-scip-query --version
-scip-query doctor
-```
-
-If the command is missing, install or link it using the repo's package manager, then rerun this step.
-
-3. Ensure bundled skills are available and project-local lifecycle hooks are configured.
-
-```bash
-scip-query install-skills
-scip-query setup-hooks --json
-```
-
-The setup command also installs skills and configures project-local hooks, but running `install-skills` directly is useful when refreshing a user-level Codex, Claude Code, or shared agents skill root. Use `setup-hooks` when the user wants to repair the current repository's reviewable `.codex/hooks.json` and `.claude/settings.json` files without rerunning the full setup pass.
-
-4. Run the robust local setup pass.
-
-```bash
-scip-query setup --json
-```
-
-This pass installs bundled skills, configures project-local hooks, checks indexer readiness, attempts configured indexer remediation, refreshes or reuses the index, verifies capabilities, smoke-tests representative command families, writes `docs/scip-query/health-dossier.md` and `docs/scip-query/health-dossier.json`, and seeds agent guidance.
-
-If setup reports `blocked`, run:
-
-```bash
-scip-query check-deps
-scip-query capability-matrix --json
-```
-
-Fix the reported missing indexer, toolchain, language, or config problem before continuing. Do not call a missing capability "clean"; record it as unavailable or blocked.
-
-5. Initialize minimal config only when the project needs one.
-
-```bash
-scip-query init
-scip-query config-validate
-```
-
-Keep `.scipquery.json` minimal. Add settings only when they are justified by observed repo facts, reviewed false positives, declared coupling, or explicit suppressions.
-
-For TypeScript monorepos with multiple discovered project shards, prefer persistent `indexer.typescript.projectMode: "workspace"`. Add `indexerConcurrency` only when the repo has enough CPU/memory or measured cold-index timing justifies a project-specific worker cap.
-
-6. Inspect freshness and refresh only when setup repairs anything or the index is not fresh.
-
-```bash
-scip-query status --capabilities
-# If freshness is stale, missing, or unknown:
-# scip-query reindex
-scip-query capabilities --json
-scip-query stats
-```
-
-If an upstream indexer skipped source files that should still be searchable or visible to agents, run:
-
-```bash
-scip-query augment-sources
-```
-
-Project hooks auto-refresh stale indexes for agent sessions unless `.scipquery.json` sets `watch.autoRefresh: false`. Use `scip-query watch` only when a human explicitly wants a long-running live watcher in a terminal.
-
-7. Confirm health signals before cleanup.
-
-Invoke `scip-health-audit` after setup when the user wants cleanup, a post-setup follow-through, or a perfect-code pass. The agent must tell the user the health score and list confirmed items before changing application code.
-
-```bash
-scip-query health --json --full
-scip-query diff-gate --json
-```
-
-8. Calibrate project config.
-
-Open or create `.scipquery.json`. Keep config minimal and tied to observed repo facts.
-
-```json
-{
-  "languages": ["typescript"],
-  "locality": {
-    "architecturalBoundarySegments": ["effect", "errors", "policies", "workflows"]
-  }
-}
-```
-
-An architectural boundary segment is a folder name that marks a code ownership boundary, such as `effect`, `errors`, `policies`, `routes`, `schemas`, or `workflows`. Its defining trait is that broad use of code inside that folder can mean the boundary is central and intentional, so `locality-candidates` should not invent a generic `shared` destination for it.
-
-Before writing `locality.architecturalBoundarySegments`, classify boundary maturity:
-
-- Mature boundary: folder names recur as intentional ownership units in code, tests, docs, route/workflow conventions, or agent standards. Add only the repo-specific segment names not already built in.
-- Emerging boundary: folder names look meaningful, but ownership is inconsistent or only visible in one area. Do not configure yet; record candidates and rerun after a few reviewed locality results prove the boundary.
-- No reliable boundary: the repo is mostly flat, mixed, generated, or ad hoc. Leave locality config absent or empty. Do not invent architecture; use `locality-candidates` as a discovery report and ask for human ownership decisions.
-
-Treat boundary maturity as an evidence question. A segment needs at least two independent signals before it belongs in config: directory structure plus a project standard, repeated import convention, test organization, package boundary, route family, workflow family, or reviewed locality false positive. If the same folder name means different things in different parts of the repo, do not add it globally.
-
-For an unstructured repo, setup runs in discovery mode. Discovery mode means the tool records candidate ownership clusters and uncertain folders, but does not write `locality.architecturalBoundarySegments`. The setup note should say which paths need a human architecture decision and should keep `locality-candidates` recommendations as suggestions to inspect, not moves to trust.
-
-Use `locality.architecturalBoundarySegments` only when validation finds a repo-specific mature boundary that is not already in the built-in list. Add folder names, not paths.
-
-9. Validate directory-organization signals when setup touches locality.
-
-```bash
-scip-query locality-candidates --json --full
-```
-
-Review rows with non-null `suggestedHome`. Keep exact destinations only when the path already exists, stays inside the source root, and matches the repo's ownership model. When a generic `shared` destination is wrong because the current folder is a real boundary, add that folder name to `.scipquery.json` and rerun.
-
-Validate both suppression and emission. Suppression means a bad generic home was withheld because the source already lives in a real boundary. Emission means a useful `suggestedHome` is still shown for code that genuinely lacks a home. A config that suppresses every row is acceptable only when manual review confirms there were no useful destinations to emit.
-
-If the repo has no mature boundaries, the correct setup result is a note, not a config guess: "locality boundaries not configured; repo needs architecture ownership decisions first." A slop codebase is a codebase whose files are arranged by accident, convenience, or recent edits rather than by stable ownership rules. Its defining trait is that directory names do not reliably predict where code should live, so the setup skill must not make `locality-candidates` appear more certain than the repo itself.
-
-10. Finish with the standard gate.
-
-```bash
-scip-query status --capabilities
-# If freshness is stale, missing, or unknown:
-# scip-query reindex
-scip-query diff-gate --json
-```
-
-Fix findings or state the explicit acceptance reason.
-
-Do not run `scip-query setup-ci` as part of this setup workflow. CI setup is intentionally deferred until that command has a mature validation path.
-
-## Suppressions
-
-Prefer fixing findings. Use suppressions only for intentional design, compatibility shims, framework entry points, or accepted false positives with a specific reason.
-
-```bash
-scip-query suppress <id> --reason "<specific reason>"
-scip-query config-validate
-```
-
-A suppression is a durable acceptance record. It should explain why the code should remain as-is, not merely silence a noisy report.
-
-## Handoff Notes
-
-When setup changes are complete, record:
-
-- repository path and revision;
-- index command and result;
-- capability matrix limitations;
-- `.scipquery.json` changes and why each setting exists;
-- validation commands and raw-output locations;
-- final `diff-gate` result.
-
-Do not add broad analyzer config in advance. Each setting should be justified by a reviewed false positive, false negative, capability gap, or repo convention.

package/skills/scip-query-setup/agents/openai.yaml

@@ -1,3 +0,0 @@
-interface:
-  display_name: "scip-query setup"
-  short_description: "Adopt or refresh scip-query indexing, agent guidance, capability checks, and project config"

package/vendor/scip/README.md

@@ -1,6 +0,0 @@
-# Managed scip binaries
-
-These binaries are built from https://github.com/scip-code/scip.git at v0.8.1.
-They are included so Windows installs can run `scip-query reindex` without requiring Go or WSL.
-
-Run `npm run build:scip-windows` before publishing to refresh them.