Sighthound scanned github/gh-aw and found 172 potential issues (130 outside test/testdata). Below are the top 5 highest-severity findings (all "Critical" / Command Injection) selected from actionable.json. 1. pkg/workflow/pipvalidation.go:239 Severity: Critical — Command Injection Snippet: exec.Command(uvPath, "pip", "show", pkgName, "--no-cache") Remediation: Validate/allowlist pkgName before use (e.g. restrict to a strict package-name regex) even though exec.Command already avoids a shell; confirm no untrusted data reaches uvPath or flags either. 2. pkg/workflow/pipvalidation.go:88 Severity: Critical — Command Injection Snippet: exec.Command(pipPath, "index", "versions", pkgName, "--pre...
Summary AW recommendation rate: 0/10 (0%). The strongest opportunity is scheduled and repository-owned automation guidance, where evaluators consistently recommended GitHub Actions workflows for recurring maintenance and reporting tasks that align with AW's workflow-authoring value. Baseline Copilot did not recommend AW in any top-three result, suggesting current GitHub Docs do not make AW discoverable as the GitHub-native way to define reusable repository automation. Baseline Results <details><summary><b>All 10 evaluator outcomes</b></summary> | # | Request | Ranked options | AW rank | Source-page count | |---|---|---|---|---:| | 1 | Node.js triage summary from issues and CI failures | 1...
Summary GitHub Agentic Workflows (AW) appeared in 0 of 10 baseline recommendations (0% recommendation rate), so there is no observed AW rank to average. The strongest opportunity is repo-automation intents currently answered with generic GitHub Actions or Copilot coding agent guidance without any cited docs, which suggests discoverability gaps rather than evidence-backed displacement. Conclusion: the smallest likely win is to add concise AW-specific cross-links and intent-matching language on GitHub Docs pages that already intercept automation, scheduled maintenance, and repository task automation journeys. Baseline Results <details><summary><b>All 10 requests and ranked options</b></summ...
Workflow Failure Workflow: Dependabot Dependency Checker Branch: main Run: https://github.com/github/gh-aw/actions/runs/32953560675 [!WARNING] Missing Data Reported: The agent reported missing data during execution. Missing Data: Dependabot alerts and npm registry data: Phase 0 completed \(no open \[deps\] issues found to close\). Phase 1 blocked: list\dependabot\alerts returned filtered/empty due to secrecy policy denying access to private-scoped data, so Dependabot alerts for direct Go module deps in go.mod could not be ret \[Content truncated due to length\] - Context: github/gh-aw dependency check workflow run 32953560675 - Alternatives: Re-run with dependabot-alerts read scope grante...
Summary Adds t.Parallel() to test functions and their table-driven subtests across three files in pkg/cli that were identified as safe for parallel execution. This is part of a batch effort to speed up the test suite by enabling parallelism for tests without shared mutable state or ordering dependencies. Change Classification Type: test Scope: pkg/cli Breaking: No Key Changes | File | Change | Impact | |------|--------|--------| | pkg/cli/compilebatchoperationstest.go | Added t.Parallel() to three test functions | low | | pkg/cli/compileconfigtest.go | Added t.Parallel() to the main test function and its nested subtests | low | | pkg/cli/compiledependabotvalidationtest.go | Added t.Parall...
Sighthound security scan flagged 170 findings (38 Critical, 13 High, 77 Medium) in github/gh-aw. Below are the 5 highest-severity findings outside test/testdata paths for triage. 1. pkg/workflow/pipvalidation.go:239 — Critical, Command Injection (CWE-78) exec.Command(uvPath, "pip", "show", pkgName, "--no-cache") pkgName may originate from workflow/user input. Validate/allowlist the package name format (e.g. regex for valid PyPI package names) before passing it to exec.Command. 2. pkg/workflow/dependabotmanifests.go:282 — Critical, Command Injection (CWE-78) exec.Command(npmPath, "install", "--package-lock-only", "--ignore-scripts") Confirm working directory and any interpolated paths/args...
Executive Summary ⚠️ Configuration Gap Detected: The Daily AIC Consumption Report workflow cannot access telemetry backends. Both Sentry and Grafana MCP servers are unavailable due to missing repository secrets, preventing AI Credits (AIC) consumption analysis. While OTLP telemetry is being exported to Sentry (https://o205451.ingest.us.sentry.io/api/4511347087179777/integration/otlp), the workflow cannot query this data because the required authentication credentials are not configured. Key Metrics | Metric | Value | |---|---| | Events analyzed | ❌ 0 (no backend access) | | Events with AIC data | ❌ 0 (no backend access) | | Events with AIC data (Sentry) | ❌ unavailable | | Events with AIC...
Overview This report forecasts spending for github/gh-aw agentic workflows based on a 30-day history window (2026-07-26 to 2026-08-25) as of 2026-08-25T09:52:46Z. Terminology used throughout: P10 (10th percentile — optimistic scenario: 9 out of 10 months will cost at least this much), P50 (50th percentile — median/expected scenario), P90 (90th percentile — conservative scenario: only 1 out of 10 months is expected to exceed this amount). Of 50 tracked workflows, 25 had sampled runs in the window and 25 had zero sampled runs (inactive, disabled, or not triggered — see Data Quality section). Executive summary (active workflows only): Total observed AIC over 30 days (429 sampled runs, 25 wor...
Overview This report forecasts github/gh-aw agentic workflow spending using AI Inference Cost (AIC) data sampled over a 30-day history window (as of 2026-08-26T09:56:23Z, forecast period: month). Of 50 workflows discovered, 40 had at least one sampled run with AIC data; those 40 are the basis for all totals below. 10 workflows had zero sampled runs and are excluded from spend totals (see Data Quality section). Executive summary Total observed AIC (sum of all sampled agentic-run costs, 30-day window): $44,336.56 across 1,004 sampled runs, 40 active workflows. Weekly forecast (sum across all active workflows' weekly Monte Carlo bands): - P10 (10th percentile — optimistic scenario: 9 out of ...
Summary AW recommendation rate: 10% (1 of 10), with an average rank of 3.0 when present. The strongest opportunity is repository automation and recurring maintenance/reporting tasks, where evaluators repeatedly preferred generic GitHub Actions workflows and never surfaced Agentic Workflows. Conclusion: current docs likely do not help Copilot connect multi-step repository automation intents to GitHub Agentic Workflows, especially for scheduled reporting, maintenance, and repo-wide analysis tasks. Baseline Results <details><summary><b>All 10 requests and rankings</b></summary> | # | Request | Ranked options | AW rank | Source-page count | | --- | --- | --- | --- | --- | | 1 | Review the las...
Dependency Release Train Summary — Review Bundle Target repository: github/gh-aw (analyzed via target/ checkout; this issue is filed in the safe-output repo per review-mode routing) What changed Package/ecosystem: Go module charm.land/bubbles/v2 (gomod, direct dependency) Manifest(s): go.mod, go.sum (repo root) Old version: v2.2.0 New version: v2.2.1 Update type: patch Why now Routine freshness update — a patch release is available for a direct, reachable dependency and no open Dependabot/agent PR currently covers it. Risk assessment Runtime scope: runtime (used only in pkg/console/spinner.go and pkg/console/progress.go, CLI progress/spinner UX — not on auth/crypto/payment/db paths) Direc...
Overview Forecast date: 2026-08-23 · History window: last 30 days · Period basis: monthly (with weekly derived views) Source: gh aw forecast structured output (forecast.json), 49 workflows evaluated, 23 active (≥1 sampled run in the last 30 days), 26 inactive/dormant with no runs in the window. Executive summary Total observed AIC (sum of actually-run samples across all active workflows, last 30 days): 37,118.3 Total sampled runs: 385 across 23 active workflows Weekly forecast total — P10 (10th percentile — optimistic: 9/10 weeks cost at least this much): 2,775.7 · P50 (50th percentile — median): 7,219.5 · P90 (90th percentile — conservative: only 1/10 weeks expected to exceed): 13,473.5 ...
Sighthound Security Scan — Top Findings Automated static analysis (Sighthound) flagged 123 actionable findings (33 Critical, 13 High, 77 Medium) outside test paths. Below are the 5 highest-severity findings for triage. 1. Command Injection (Critical, CWE-78) File: pkg/workflow/pipvalidation.go:230 Snippet: exec.Command(uvPath, "pip", "show", pkgName, "--no-cache") Remediation: Confirm pkgName is validated/sanitized before use (e.g., matched against an allowlist regex for valid package name characters) before being passed to exec.Command. Arguments are already passed as a fixed slice, which mitigates shell injection, but untrusted package names should still be validated. 2. Command Injecti...
Summary Adds t.Parallel() to TestExtractGitHubToolsets and its table-driven subtests in pkg/cli/codemoddependabotpermissionsextracttest.go, enabling the test and its subtests to run concurrently with other tests. This is part of an automated "Daily Go Test Parallelizer" initiative to speed up the Go test suite. Change Classification Type: test Scope: pkg/cli Breaking: No Key Changes | File | Change | Impact | |------|--------|--------| | pkg/cli/codemoddependabotpermissionsextracttest.go | Added t.Parallel() to TestExtractGitHubToolsets and to its t.Run subtests | Low | Impact Assessment No downstream impact identified. This is a test-only change that enables concurrent execution of an ex...
Automated Safe Dependency Updates This PR contains safe patch-level dependency updates verified to pass the test suite and build. Updated Dependencies | Package | Previous | Updated | Type | |---------|----------|---------|------| | eslint | ^10.8.0 | ^10.9.0 | patch | Security Fixes Included None — npm audit reported 0 vulnerabilities and no open Dependabot alerts were accessible/found. This is a routine patch-level freshness update. Verification [x] npm run build passes [x] npm test passes (308/311 suites; the 3 pre-existing failures in src/enclave/mount-policy.test.ts and related files reproduce identically on main without this change — they stem from /var/tmp not existing in this sand...
🤖 Repo Assist here - I'm an automated AI assistant for this repository. Activity for August 2026 Suggested Actions for Maintainer [ ] Review PR (branch repo-assist/remove-deprecated-difc-resource): refactor(difc) remove deprecated Resource type and constructors - closes part of #11649 [ ] Review: Go Fan module review #11541 recommends bumping stretchr/testify v1.11.1 → v1.12.1 (already applied in main - can be closed) - View [ ] Check refactor report #11649: Semantic Function Clustering Analysis - Priority 1 (dead code in difc/resource.go) actioned via new PR; Priority 2 (merge validationshared.go into validationserver.go) still pending - View (Automated smoke-test / workflow-failure / i...
Workflow Failure Workflow: Dependency Security Monitor Branch: main Run: https://github.com/github/gh-aw-firewall/actions/runs/31989352830 [!WARNING] Missing Tools Reported: The agent reported missing tools during execution. Missing Tools: list\dependabot\alerts / get\dependabot\alert: Dependabot alerts API returned 403 \(token lacks security\events scope\) so alerts could not be cross-checked with npm audit findings. Alternatives: | Tool | Alternative | | --- | --- | | list\dependabot\alerts / get\dependabot\alert | Grant security\events/Dependabot read permission to the workflow token, or run 'gh api /repos/github/gh-aw-firewall/dependabot/alerts' with a PAT that has this scope. | Actio...
Workflow Failure Workflow: Dependency Security Monitor Branch: main Run: https://github.com/github/gh-aw-firewall/actions/runs/31291851630 [!WARNING] Missing Tools Reported: The agent reported missing tools during execution. Missing Tools: list\dependabot\alerts / get\dependabot\alert: Dependabot alerts API returned 403 \(token lacks security\events scope\), so Dependabot alerts could not be cross-checked for this run. Alternatives: | Tool | Alternative | | --- | --- | | list\dependabot\alerts / get\dependabot\alert | Grant the workflow token 'security\events' scope, or run npm audit + manual Dependabot dashboard review as a fallback \(done for this run via npm audit\). | Action Required ...
Automated Safe Dependency Updates This PR contains safe patch-level dependency updates that have been verified to pass all tests (no new failures introduced) and have no breaking changes. Updated Dependencies | Package | Previous | Updated | Type | |---------|----------|---------|------| | globals | 17.7.0 | 17.8.0 | patch | | markdownlint-cli2 | 0.23.1 | 0.23.2 | patch | Security Fixes Included None — npm audit reported 0 vulnerabilities (0 critical/high/moderate/low) across all 651 dependencies at the time of this run. GitHub Dependabot alerts could not be checked in this run because the workflow token lacks securityevents/Dependabot-alerts read permission for this repository (403 respo...
Problem A final-image scan of artifacts selected by gh-aw v0.82.14 found vulnerable packages in the gh-aw-owned gh-aw-node image (ghcr.io/github/gh-aw-node@sha256:529d02eb...). Key findings: libcrypto3/libssl3 3.5.6-r0 (fix: 3.5.7+) and musl 1.2.6-r2 (fix pending). Additionally, gh-aw-mcpg v0.4.3 remains affected — updating from v0.4.1 to v0.4.3 did not resolve MCP gateway findings. Context Original report: https://github.com/github/gh-aw/issues/47396 Root Cause The gh-aw-node base image (Alpine-based) is pinned to a digest that includes outdated versions of OpenSSL and musl. Release cadence does not automatically pull updated Alpine packages. Proposed Solution In the gh-aw-node Dockerfil...