SocketDev/action - Page 2

optimizing-parser-performance

Audits parser and tokenizer performance.

optimizing-performance

Audits performance in Rust, Go, C++, JS/TS, parsers, Node addons, and WebAssembly.

optimizing-react-interfaces

Optimizes measured React interface performance.

optimizing-rust-performance

Audits measured Rust performance.

optimizing-submodules

Minimize submodule sparse-checkouts to only the trees the repo actually consumes.

optimizing-webassembly-performance

Audits WebAssembly performance and capability fallbacks.

patching-findings

Fix verified security findings with minimal patches and independent review before validation.

plugging-promise-race

Reference for avoiding Promise.race/any handler leaks in loops and hand-rolled concurrency pools.

property-and-fuzz-testing

Pick a property/fuzz tier and harness for parsers, decoders, and native addons.

prose

Edit human-facing prose to remove AI patterns, padding, hedging, and awkward generated phrasing.

pushing

Run the full pre-push gate, push only when green, then watch CI after the push.

refreshing-history

Refresh a default branch to one signed initial commit: update deps, verify, force-push with backup.

releasing-a-package

Release a single-package fleet repo: pre-bump, changelog, staged publish, approve, tag, release.

rendering-chromium-to-png

Render pages or extension popups to PNG and inspect pixels to catch real UI regressions.

reordering-release-bump

Move an existing version-bump commit back to tip, retag it, verify integrity, and force-push safely.

researching-recency

Research recent community and upstream signals across GitHub, forums, social feeds, and web sources.

reviewing-code

Review the current branch against a base with multiple AI backends and produce verified findings.

reviewing-web-interfaces

Reviews web interface quality before landing UI changes.

running-test262

Run the repo's canonical Test262 suite for parser/runtime conformance; never write ad hoc runners.

scanning-quality

Scan for bugs, races, workflow flaws, insecure defaults, and regressions; verify and rank findings.

scanning-security

Run AgentShield, zizmor, and Socket dependency scans into a graded security report.

scanning-vulns

Run static vulnerability scanners over a target tree and write raw findings for triage.

setup-repo

Repo onboarding wizard: tokens, keychain, shell bridge, tools, hooks, initialization.

squashing-history

Squash default-branch history to one commit with backup and force-push.

testing-web-interfaces

Tests web interfaces across component and browser layers.

threat-modeling

Build or interview for a threat model covering assets, attackers, trust boundaries, and mitigations.

tidying-files

Remove untracked or ignored junk and scratch files across fleet repos without touching tracked work.

tidying-rolldown-bundles

Keep rolldown bundles lean by reporting collapsible deps, catalog drift, and fat external shims.

tidying-worktrees

Sweep spent clean worktrees whose branches merged or vanished; preserve dirty or unpushed work.

triaging-findings

Verify raw security findings, dedupe them, rerank exploitability, and assign owners.

trimming-bundle

Find unused bundled code paths and keep only stubs that rebuild and pass tests.

updating-coverage

Refresh the README coverage badge by running coverage and rewriting the shields.io badge line.

updating-daily

Promote soak-cleared dependency exclusions, drop expired dates, and reconcile the lockfile.

updating-hooks-dry

Read-only DRY sweep of fleet hooks and oxlint rules for overlap and consolidation.

updating-lockstep

Resolve lockstep manifest drift by pipelining version-pin rows and reporting parity rows.

updating-pricing

Refresh current AI model pricing data from vendor sources and update fleet pricing constants.

updating-security

Resolve Dependabot alerts by bumping, overriding, patching, or dismissing with evidence.

updating

Run repo maintenance: updates, lockstep, submodules, security, coverage, audits.

writing-disclosures

Write or review a dual-use DISCLOSURE file; npm Trust & Safety reads it, so every claim must be verifiable.

writing-fast-tests

Writing or reviewing tests, or a slow suite: pick the cheapest seam, keep files parallel-safe.