@rune-kit/rune 2.2.6 → 2.3.0

package/skills/review/SKILL.md

@@ -3,7 +3,7 @@ name: review
 description: Code quality review — patterns, security, performance, correctness. Finds bugs, suggests improvements, triggers fix for issues found. Escalates to opus for security-critical code.
 metadata:
   author: runedev
-  version: "0.3.0"
+  version: "0.5.0"
   layer: L2
   model: sonnet
   group: development
@@ -70,6 +70,28 @@ Determine what to review.
 - If context is unclear: use `rune:scout` to identify all files touched by the change
 - List every file in scope before proceeding — do not review files outside the stated scope
 
+### Step 1.5: Blast Radius Assessment
+
+For each modified function/class, estimate its blast radius before reviewing.
+
+```
+Use Grep to count direct callers/importers of each modified symbol:
+  blast_radius = count(files importing or calling this symbol)
+```
+
+| Blast Radius | Risk | Review Depth |
+|-------------|------|-------------|
+| 1-5 callers | Low | Standard review |
+| 6-20 callers | Medium | Check all callers for compatibility |
+| 21-50 callers | High | Thorough review + regression test check |
+| 50+ callers | Critical | MUST escalate to adversarial analysis (rune:adversary) even in quick triage |
+
+<HARD-GATE>
+Modifying a symbol with 50+ callers + HIGH severity change (logic, types, behavior) → adversarial analysis REQUIRED. Quick review is NOT sufficient for high-blast-radius changes.
+</HARD-GATE>
+
+> Source: trailofbits/skills (3.7k★) — quantitative blast radius as escalation threshold.
+
 ### Step 2: Logic Check (Production-Critical Focus)
 
 Read each changed file. Prioritize bugs that **pass CI but break production** — these are the highest-value findings because linters and type checkers already catch the rest.
@@ -156,6 +178,24 @@ Check for security-relevant issues.
 - If any security-sensitive code found (auth logic, input handling, crypto, payment): call `rune:sentinel` for deep scan
 - Sentinel escalation is mandatory — do not skip it for auth or crypto code
 
+### Step 4.5: API Pit-of-Success Check
+
+For code that exposes APIs, shared utilities, or reusable interfaces, evaluate through 3 adversary personas:
+
+| Adversary | Mindset | What They Reveal |
+|-----------|---------|-----------------|
+| **The Scoundrel** | Malicious — controls config, crafts inputs, exploits edge cases | Security holes, privilege escalation, injection surfaces |
+| **The Lazy Developer** | Copy-pastes from docs, skips error handling, uses defaults | Unsafe defaults, missing validation, footgun APIs |
+| **The Confused Developer** | Misunderstands API semantics, passes wrong types, ignores return values | Ambiguous interfaces, poor naming, missing type safety |
+
+**Pit-of-Success principle**: Secure, correct usage should be the path of least resistance. If the API makes it EASIER to use it wrong than right → WARN.
+
+Check: Does the API have sensible defaults? Does misuse fail loudly (not silently)? Is the happy path obvious from the signature?
+
+**Skip if**: Code is internal-only (no external consumers), single-use utility, or test-only.
+
+> Source: trailofbits/skills (3.7k★) — 3 adversary types for API review.
+
 ### Step 5: Test Coverage
 
 Identify gaps in test coverage.
@@ -324,6 +364,36 @@ className="focus:outline-none focus-visible:ring-2 focus-visible:ring-blue-500"
 - iOS target: solid-background cards (iOS 26 Liquid Glass deprecates this visual language) — should use translucent/blur surfaces
 - Android target: hardcoded hex colors instead of `MaterialTheme.colorScheme` tokens → not adaptive to dynamic color
 
+## Weighted Composite Scoring
+
+When a review is part of a recurring quality-gate cycle (e.g., sprint review, pre-release gate), produce a **composite quality score** alongside the findings list. This makes review output numeric and comparable across runs.
+
+### Formula
+
+```
+Quality Score = (Correctness × 0.35) + (Security × 0.30) + (Test Coverage × 0.20) + (Conventions × 0.15)
+```
+
+Each dimension is scored 0–100 based on findings count and severity:
+- 0 CRITICAL/HIGH findings → 100 for that dimension
+- 1 CRITICAL → dimension capped at 40
+- 1 HIGH → dimension capped at 70
+- Each additional MEDIUM → subtract 5 (floor: 50)
+
+### Grade Thresholds
+
+| Score | Grade | Verdict |
+|-------|-------|---------|
+| 90–100 | Excellent | APPROVE |
+| 75–89 | Good | APPROVE with notes |
+| 60–74 | Fair | REQUEST CHANGES (MEDIUM issues) |
+| 40–59 | Poor | REQUEST CHANGES (HIGH issues present) |
+| 0–39 | Critical | REQUEST CHANGES (CRITICAL present) |
+
+**When to include**: Only when `mode: "scored"` is passed by the caller, or when invoked by `audit`. Default review output uses the standard severity-ranked report without the score.
+
+> Source: zubair-trabzada/geo-seo-claude (2.7k★) — weighted composite formula, 5-tier grade thresholds.
+
 ## Severity Levels
 
 ```
@@ -339,8 +409,12 @@ LOW       — style inconsistency, naming suggestion, minor refactor opportunity
 ## Code Review Report
 - **Files Reviewed**: [count]
 - **Findings**: [count by severity]
+- **Review Commit**: [git hash at time of review]
 - **Overall**: APPROVE | REQUEST CHANGES | NEEDS DISCUSSION
 
+### Spec Compliance
+- [PASS/FAIL]: [acceptance criteria coverage]
+
 ### CRITICAL
 - `path/to/file.ts:42` — [description of critical issue]
 
@@ -350,6 +424,9 @@ LOW       — style inconsistency, naming suggestion, minor refactor opportunity
 ### MEDIUM
 - `path/to/file.ts:120` — [description of medium issue]
 
+### Blast Radius
+- [High-impact symbols with caller counts]
+
 ### Positive Notes
 - [good patterns observed]
 
@@ -357,6 +434,18 @@ LOW       — style inconsistency, naming suggestion, minor refactor opportunity
 [Summary and recommendation]
 ```
 
+### Review Staleness Detection
+
+Track the git commit hash at review time. If code changes after review → review is STALE.
+
+```
+Review commit: abc123 → Code changed to def456 → Review is STALE, re-review required
+```
+
+When `cook` or `ship` checks review status: compare review commit hash with current HEAD. If different → WARN: "Review is stale — code changed since last review."
+
+> Source: garrytan/gstack v0.9.4 (50.9k★) — commit hash staleness tracking for reviews.
+
 ## Constraints
 
 1. MUST read the full diff — not just the files the user pointed at
@@ -367,6 +456,16 @@ LOW       — style inconsistency, naming suggestion, minor refactor opportunity
 6. MUST escalate to sentinel if auth/crypto/secrets code is touched
 7. MUST flag untested code paths and recommend tests via rune:test
 
+## Returns
+
+| Artifact | Format | Location |
+|----------|--------|----------|
+| Code review report | Markdown | inline (chat output) |
+| Severity-ranked findings | Markdown table | inline |
+| Spec compliance verdict | Markdown | inline |
+| Composite quality score | Markdown table | inline (when `mode: "scored"`) |
+| Blast radius assessment | Markdown table | inline |
+
 ## Sharp Edges
 
 | Failure Mode | Severity | Mitigation |

package/skills/review-intake/SKILL.md

@@ -233,6 +233,17 @@ How to push back:
 - Tests pass after each individual fix
 - Review Intake Report emitted
 
+## Returns
+
+| Artifact | Format | Location |
+|----------|--------|----------|
+| Review Intake Report | Markdown table | inline |
+| Categorized feedback (P0–P4) | Classified list | inline |
+| Verdict per item (CORRECT/PUSHBACK/YAGNI/DEFER) | Table | inline |
+| Action plan (changes applied) | File list with descriptions | inline |
+
 ## Cost Profile
 
 ~2000-5000 tokens depending on feedback volume. Sonnet for evaluation logic, haiku for scout/grep verification.
+
+**Scope guardrail:** review-intake processes the feedback items provided — it does not pull new reviews, open PRs, or change architectural decisions without owner confirmation.

package/skills/safeguard/SKILL.md

@@ -183,6 +183,18 @@ Known failure modes for this skill. Check these before declaring done.
 - Git tag `rune-safeguard-<module>` created
 - Safeguard Report emitted with test count, coverage, and rollback tag
 
+## Returns
+
+| Artifact | Format | Location |
+|----------|--------|----------|
+| Characterization test file | TypeScript/JS/Python test | `tests/char/<module>.test.*` |
+| Boundary markers | Code comments (@legacy, @bridge) | in-source |
+| Frozen config snapshot | Copies of config files | `.rune/*.frozen.*` |
+| Git rollback tag | Git tag | `rune-safeguard-<module>` |
+| Safeguard Report | Markdown | inline |
+
 ## Cost Profile
 
 ~2000-5000 tokens input, ~1000-2000 tokens output. Sonnet for test writing quality.
+
+**Scope guardrail:** safeguard builds safety nets only — it does not refactor code. All surgery is delegated to `surgeon` after the safeguard HARD-GATE passes.

package/skills/scaffold/SKILL.md

@@ -247,6 +247,16 @@ Max 3 fix-verify loops. If still failing after 3 → report failures to user wit
 9. MUST respect user's existing project if scaffolding into non-empty directory — warn and ask before overwriting
 10. Generated files MUST be < 500 LOC each — split large files
 
+## Returns
+
+| Artifact | Format | Location |
+|----------|--------|----------|
+| Project directory structure | Directories + files | Project root (per plan) |
+| Source code | Source files | Per plan file list |
+| Test suite | Source files | Co-located or `tests/` per framework convention |
+| Documentation | Markdown | `README.md`, `ARCHITECTURE.md`, `docs/API.md` as applicable |
+| Scaffold Report | Markdown (inline) | Emitted at session end |
+
 ## Sharp Edges
 
 | Failure Mode | Severity | Mitigation |

package/skills/scope-guard/SKILL.md

@@ -145,6 +145,17 @@ Known failure modes for this skill. Check these before declaring done.
 - Creep severity classified (IN_SCOPE / MINOR CREEP / SIGNIFICANT CREEP)
 - Scope Report emitted with recommendations
 
+## Returns
+
+| Artifact | Format | Location |
+|----------|--------|----------|
+| Scope Report | Markdown (IN_SCOPE / MINOR CREEP / SIGNIFICANT CREEP) | inline |
+| In-scope file list | Classified list | inline |
+| Out-of-scope drift report | File list with reasons | inline |
+| Recommendations | Actionable list | inline |
+
 ## Cost Profile
 
 ~200-500 tokens input, ~100-300 tokens output. Haiku. Lightweight monitor.
+
+**Scope guardrail:** scope-guard reports drift and advises — it does not revert files, block commits, or modify code. Override decisions belong to the calling orchestrator or the user.

package/skills/scout/SKILL.md

@@ -206,6 +206,13 @@ None — pure scanner using Glob, Grep, Read, and Bash tools directly. Does not
 - [pattern or potential issue noticed]
 ```
 
+## Returns
+
+| Artifact | Format | Location |
+|----------|--------|----------|
+| Scout Report | Markdown (inline) | Emitted to calling skill |
+| Codebase map | Markdown | `.rune/codebase-map.md` (when called by cook, team, onboard, autopsy) |
+
 ## Sharp Edges
 
 Known failure modes for this skill. Check these before declaring done.
@@ -230,3 +237,5 @@ Known failure modes for this skill. Check these before declaring done.
 ## Cost Profile
 
 ~500-2000 tokens input, ~200-500 tokens output. Always haiku. Cheapest skill in the mesh.
+
+**Scope guardrail**: Do not expand the scan to unrelated modules or write files beyond `.rune/codebase-map.md` unless explicitly delegated by the parent agent.