freeze-mode
Use when you need a read-only investigation window for production or sensitive systems and want Claude blocked from non-read-only tools plus non-read-only…
Skill by leo110047
Chalk Labs
Need help deploying freeze-mode?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/leo110047-freeze-mode
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: agent-building
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
- Runs as
- Node.js package
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/freeze-mode
What's inside
- Skill name
freeze-mode- Description
- Use when you need a read-only investigation window for production or sensitive systems and want Claude blocked from non-read-only tools plus non-read-only Bash commands. Best fit for incident triage, audits, or "look but do not change" sessions.
- License
- MIT
- Instructions
- 52 lines in SKILL.md · view on GitHub
Before installing
- Ships executable scripts
Agent Skills spec issues
- allowed-tools: must be a space-separated string
Files (2)
- SKILL.md
- freeze-mode.js · scripts/
SKILL.md (first lines)
# Freeze Mode
## When to Use
- Production debugging, audits, incident review, or evidence collection that
must not mutate anything.
- Work that stays read-only until a human lifts the freeze.
- Situations where `careful-mode` guards too few operations.
## Script
- `node scripts/freeze-mode.js enable`: activates `freeze-mode` for the current
Claude session.
- `node scripts/freeze-mode.js status`: reports activation, state location, and
protections.
- `node scripts/freeze-mode.js disable`: deactivates the mode.
Flags: `--session <id>` targets another session; `--json` prints
machine-readable output.
## Protections
While active, the `PreToolUse` router:
- Blocks every non-Bash tool outside `READ_ONLY_TOOLS` in
`hooks/scripts/lib/hook-router/pretool-policy.js`.
- Blocks every Bash command outside `FREEZE_MODE_ALLOWED_BASH` in
`hooks/scripts/lib/hook-router/freeze-mode-rules.js`.
- Blocks shell chaining, command substitution, pipes, and redirections.
## Gotchas
- The mode adds friction; it does not replace change control, approvals, or
runbooks.
- The allowlist favors false positives: command families with mutating forms
are excluded entirely.
- Disable the mode before implementing fixes.
- Mode state is session-scoped; check `status` before claiming a session stayed
read-only.
- Without `${CLAUDE_PLUGIN_DATA}`, mode state falls back to temp storage.
## Completion Check
Before claiming the session stayed read-only, report:
- The command used to enable the mode.
- The `status` evidence before the investigation.
- The read-only evidence gathered.
- Any blocked tool or shell attempt.
- The `status` evidence after disabling the mode.From leo110047/goldband, MIT.
More from leo110047/goldband
- 0.566score
claude-config-verification
leo110047
Use when modifying the goldband config repo across Claude Code or Codex assets (`skills/`, `hooks/`, `commands/`, `rules/`, `claude/`, `.claude-plugin/`…
SkillAFreshness: AClaude Code +9 - 0.563score
goldband
leo110047
Route Goldband tasks through the formal capability interface for review, investigation, QA, planning, browser, design, safety, context, knowledge…
SkillAFreshness: AClaude Code +9 - 0.563score
careful-mode
leo110047
Use before running destructive or prod-facing CLI commands.
SkillAFreshness: AClaude Code +9 - 0.560score
subagent-development
leo110047
Use when a task can be isolated into a fresh-context subagent or parallelized safely, and its result needs review against the original requirements.
SkillAFreshness: AClaude Code +9 - 0.560score
security-checklist
leo110047
Use when working on authentication, authorization, input validation, secret handling, sensitive data storage, API hardening, deployment security, or security…
SkillAFreshness: AClaude Code +9 - 0.560score
ci-cd-integration
leo110047
Use when setting up or changing CI pipelines, GitHub Actions workflows, build/test/deploy gates, release automation, CI secrets handling, or build caching.
SkillAFreshness: AClaude Code +9 - 0.560score
planning-workflow
leo110047
Use when turning a request into an implementation plan, especially before multi-file changes, refactors, or work that needs explicit phases and verification.
SkillAFreshness: AClaude Code +9 - 0.560score
frontend-design
leo110047
Use when creating, restyling, or reviewing frontend pages, components, dashboards, tools, posters, artifacts, or application screens.
SkillAFreshness: AClaude Code +9 - 0.560score
evidence-based-coding
leo110047
Ground code, configuration, and completion claims in evidence for the current candidate and relevant environment.
SkillAFreshness: AClaude Code +9 - 0.560score
performance-optimization
leo110047
Use for slowness or memory pressure, performance claims, and changes to bulk processing, materialization, caching, or parallel execution where workload growth…
SkillAFreshness: AClaude Code +9 - 0.560score
prompt-hygiene
leo110047
Use when creating or editing prompts, system prompts, agent instructions, workflow prompts, cron prompts, briefing prompts, design-tool prompts…
SkillAFreshness: AClaude Code +9 - 0.556score
skill-developer
leo110047
Use when creating, merging, or editing portable skills, trigger descriptions, progressive-disclosure layouts, or hook-backed skill tooling in this repo.
SkillAFreshness: AClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 5
- Last commit (this folder)
- Oct 14, 2026 (17 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/leo110047/goldband
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from leo110047
- 0.566score
claude-config-verification
leo110047
Use when modifying the goldband config repo across Claude Code or Codex assets (`skills/`, `hooks/`, `commands/`, `rules/`, `claude/`, `.claude-plugin/`…
SkillAFreshness: AClaude Code +9 - 0.563score
careful-mode
leo110047
Use before running destructive or prod-facing CLI commands.
SkillAFreshness: AClaude Code +9 - 0.563score
goldband
leo110047
Route Goldband tasks through the formal capability interface for review, investigation, QA, planning, browser, design, safety, context, knowledge…
SkillAFreshness: AClaude Code +9 - 0.560score
subagent-development
leo110047
Use when a task can be isolated into a fresh-context subagent or parallelized safely, and its result needs review against the original requirements.
SkillAFreshness: AClaude Code +9 - 0.560score
ci-cd-integration
leo110047
Use when setting up or changing CI pipelines, GitHub Actions workflows, build/test/deploy gates, release automation, CI secrets handling, or build caching.
SkillAFreshness: AClaude Code +9 - 0.560score
planning-workflow
leo110047
Use when turning a request into an implementation plan, especially before multi-file changes, refactors, or work that needs explicit phases and verification.
SkillAFreshness: AClaude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get freeze-mode running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
freeze-mode skill. Agent Mart. https://agentmart.dev/skills/leo110047-freeze-mode (last updated 2026-10-24)