Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/fold-run/fold/console-syncnpx skills add fold-run/fold --skill console-syncgit clone --depth 1 https://github.com/fold-run/foldWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00062 | $0.01286 |
| Opus 5 | $0.00031 | $0.00643 |
| Sonnet 5 | $0.00012 | $0.00257 |
| Haiku 4.5 | $0.00006 | $0.00129 |
Grade A, and why
console-sync scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.
The vendored console
gateway/console is build output from another repo, checked in.
scripts/sync-console.sh vendors it from fold-run/fold-console at a
pinned commit; gateway/console_source.go records that pin and is
generated, never hand-edited.
Assets are checked in rather than fetched because the Go module proxy is
fold's distribution channel: go run github.com/fold-run/fold/cmd/fold@latest
builds from the proxy zip alone, which runs no generators and carries no
submodule content. An unvendored tree is a //go:embed build failure for
every user installing that way.
Review a bump as a supply-chain change. These assets execute in an operator's browser, same-origin with a page holding a live Bearer token in memory. That is why nothing here auto-merges.
Two different things are called console-sync
| What it is | When it runs | |
|---|---|---|
make console-check |
CI gate proving gateway/console still equals its pin — re-vendors and fails on any diff. Its own job, because it needs network. |
Every merge |
.github/workflows/console-sync.yml |
Weekly job that resolves fold-console main and opens a bump PR. Never merges. |
Mondays 07:00 UTC |
console-check going red means someone hand-edited vendored assets, or
the pin and the tree disagree. The fix is make sync-console and a review
of what moved — never editing gateway/console directly, which is the one
thing the gate exists to catch.
Reviewing a bump PR
- Read the upstream diff, not this one. The diff here is minified
bundle output; reading it is not a review. Read
src/on the upstream commit range. Upstream CI rebuilds and fails on drift betweensrc/anddist/, so the bytes vendored here are entailed by that diff. - Check the manifest.
MANIFESTinscripts/sync-console.shis the exact allowlist of files that may enter the binary —//go:embedtakes the whole directory, so without it anything upstream's build emitted alongside (a source map, a second chunk, a stats file) would ship to every operator and be served under/console/. A bump that adds a file to the embedded set is a reviewed change here, and is also the shape a supply-chain injection would take.gateway/introspection_test.goasserts the same set from the other side, against the embedded FS. fonts/OFL.txtis not decoration. OFL 1.1 requires the licence accompany the Font Software, and these subsets are in every fold binary. It ships or the binary is out of compliance.- An empty check list means "not run", never "passed". GitHub creates
no workflow runs for branches pushed with
GITHUB_TOKEN, soci.ymlnever fires on a bot-opened bump PR. The workflow runs the manifest and API-path tests inline before opening it; push any commit to the branch to trigger the full suite before merging. - Run the
security-auditoragent over it before approving.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 88 lines · 62 tokens per session scan A 60e7ab8138c3
console-sync is a skill published in the GitHub repository fold-run/fold (1 stars, last pushed 2d ago), licensed Apache-2.0. It adds 62 tokens to every session and 1,286 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
ci-failure-triage
Use when make ci-local fails locally or a GitHub Actions run reports a CI failure (format, lint, line-budget, typecheck, or test errors).
sq-site-dependabot
Reviews, validates, and safely merges Dependabot pull requests for the sq.io site (site/, Bun lockfile). Use when clearing site dependency PRs, triaging Dependabot failures, or checking Lighthouse impact before merge.
sq-actions-dependabot
Reviews and merges Dependabot pull requests for GitHub Actions (the github-actions ecosystem) that bump uses: pins in .github/workflows/. Use for Dependabot githubactions PRs (branches like dependabot/githubactions/...), not go.mod or site/ Bun PRs.
engflow-artifacts
Use when downloading test logs, artifacts, or outputs.zip from EngFlow build invocations. Use when investigating CockroachDB CI test failures hosted on mesolite.cluster.engflow.com.
ci-orchestrator
Run a CI-like pipeline locally (format, lint, vet, static-analysis, tests) and summarize per-step results with remediation guidance.
hotplex-release
通过判断 SemVer 影响、整理 CHANGELOG、对账版本面和验证发布输入来准备或发布 HotPlex 版本。不要用于主机二进制更新、运行时诊断或文档巡逻。.