PAGES
← all tools
◆ Compliance Automation & Trust Management

GrantGuard

v0.1.0 open-source

GrantGuard audits, reviews, and removes Claude Code standing permissions, classifying allow rules into risk groups via a local web UI or CLI.

source ↗ rss ↗ What is a cyber toolchain? last repo push Jul 13, 2026

Summary

GrantGuard is an open-source, MIT-licensed command-line tool that audits the permission allowlist Claude Code builds up as users click "always allow," flagging risky grants like inline API keys, credential-store reads, destructive wildcards, and unprompted remote pushes. It runs locally as a small Python app with no third-party runtime dependencies, offering both a CLI for scanning settings files and a local browser UI, bound to 127.0.0.1, for reviewing and removing flagged entries with one click. It's aimed at developers and teams using Claude Code who need to keep standing permissions in check as part of their broader compliance posture. First committed in 2026, it has four contributors and a release as recently as 45 days ago.

GrantGuard audits, reviews, and removes Claude Code standing permissions, classifying allow rules into risk groups via a local web UI or CLI.

How GrantGuard helps

Which of my Claude Code "always allow" grants have quietly become a security risk?

scans every settings file it finds and classifies each allow rule by risk group — inline credentials, credential-store reads, destructive wildcards, overbroad wildcards, unprompted remote pushes

Is there a pasted API key or secret sitting in my standing permissions?

flags allow rules containing inline credentials and redacts secrets in its own output, so the exposure is visible without being repeated back in the clear

Which repos on this machine have permission grants nobody has reviewed?

shallow-scans a directory for settings files, or deep-scans recursively to catch ones buried in subfolders, and reports every file it audited

Can I get rid of a risky grant without losing everything else I approved?

removes only the flagged rules, one click or one non-interactive pass, and leaves the rest of the settings file untouched

What happens the next time Claude wants to do the thing I just revoked?

removals are non-destructive, so Claude simply re-asks for that permission and you decide again rather than being silently blocked

How strict should "overbroad" wildcard matching be for my team?

a tolerance setting chooses between flagging any broad wildcard by default or only the permissive ones, changing what counts as risky without changing the scan itself

all 5 features, with the evidence for each →

Languages

Not documented

Features

5 capabilities

Built from GrantGuard's releases, docs, README and our own runs. “Seen” marks the month we first spotted a release the vendor didn't date. Group names and tags are our labels.

Compare with peers →

  • GrantGuard covers 1 of the 11 endpoint investigation and controls capabilities we track.

What GrantGuard offers you

  • Command-line interface2
All capabilities 5 capabilities
Claude Code permission auditing shipped Scans Claude Code settings files across standard and project-level locations, classifies every allow rule by risk category (inline credentials, credential-store reads, destructive wildcards, unprompted remote pushes, overbroad wildcards), and reports flagged and safe rules. 4 releases · 5 other sources · first seen Jul 2026

Shared capability: Security baseline assessment

Also counts toward: Command-line interface

release

  • Reads all standard Claude Code permission sources by default: ~/.claude/settings.json, ~/.claude/settings.local.json, ~/.claude.json, and the platform managed settings file. v0.1.0 · Jul 2026 · source · release history
  • New --tolerance flag (default or permissive) controls whether overbroad wildcard rules are flagged or treated as safe. v0.1.0 · Jul 2026 · source · release history
  • New audit command scans Claude Code settings files and classifies every allow rule by risk — inline credentials, credential-store reads, destructive wildcards, unprompted remote pushes, and overbroad wildcards. v0.1.0 · Jul 2026 · source · release history
  • New --show-safe flag on audit prints kept (non-flagged) rules alongside flagged ones for full visibility. v0.1.0 · Jul 2026 · source · release history

docs

  • Audits, reviews, and removes risky Claude Code standing permissions from settings files, classifying allow rules by risk level.

example

  • Quickly audit your personal Claude Code permissions for high-risk grants — inline API keys, credential-store reads, destructive wildcards — without modifying anything. uv run grantguard.py audit v0.1.0 · Jul 2026 · source
  • Audit default Claude Code settings grantguard audit
  • Audit a specific repo's settings grantguard audit /path/to/repo
  • Audit multiple target repos grantguard audit --targets /repo/a --targets /repo/b
Settings file discovery shipped Locates Claude Code settings files via shallow directory scanning or deep recursive search under specified target directories. 1 release · 2 other sources · first seen Jul 2026

Also counts toward: Command-line interface

release

  • Supports --scan for shallow discovery of .claude/settings*.json under target directories, and --deep-scan for broad recursive discovery. v0.1.0 · Jul 2026 · source · release history

example

  • Shallow scan workspace for settings grantguard audit --scan --targets /path/to/workspace
  • Deep scan all discoverable settings grantguard audit --deep-scan
Automated permission rule removal shipped suggested Applies non-interactive fixes to flagged permission rules via a --fix flag, rewriting only the offending entries while leaving all other settings intact. 1 release · 1 other source · first seen Jul 2026

Suggested capability: Security baseline enforcement. This is our guess, not confirmed.

release

  • New --fix flag on audit applies removals non-interactively, rewriting only the flagged rules while leaving all other settings intact. v0.1.0 · Jul 2026 · source · release history

example

  • Audit and apply fixes automatically grantguard audit --fix
Interactive browser UI for permission review shipped suggested Opens a local browser interface that lets users review flagged permission grants and remove them one at a time across a repo's Claude settings. 1 release · 1 other source · first seen Jul 2026

Suggested capability: Data access governance. This is our guess, not confirmed.

release

  • New ui command opens a local browser UI for interactive review and one-click removal of flagged permission grants. v0.1.0 · Jul 2026 · source · release history

example

  • Open the interactive browser UI to review and selectively remove flagged permissions across a specific repo's Claude settings. uv run grantguard.py ui /path/to/repo v0.1.0 · Jul 2026 · source
Cross-platform support shipped Runs on macOS, Linux, and Windows with no third-party runtime dependencies. 1 release · first seen Jul 2026

release

  • Cross-platform support for macOS, Linux, and Windows with zero third-party runtime dependencies. v0.1.0 · Jul 2026 · source · release history

Lines in monospace are the tool's own words: help text parsed from its source, or an endpoint from its API document. Under release, a line is our summary of a dated release. Under docs, it is quoted from the tool's own documentation or website. Each links back to the source it came from.

Score card

  1. Shipping velocity releases a month, last 12 months
  2. Documentation quality release-notes rubric, 0–100
  3. Breaking-change discipline how rarely a release breaks users, and notice given
  4. Maintenance risk quiet time against its own cadence, maintainers
  5. Innovation & surface new capabilities, API and MCP surface
  6. Community health
  7. Contributor trend distinct committers, this year against last
  8. Maintainer concentration bus factor now against a year ago
  9. Issue backlog trend issues closed per issue opened, 12 months
  10. Responsiveness median time to a maintainer’s first reply

Nine scores for GrantGuard, each built from the releases, notes and API history this site tracks — and, on Researcher and Business, where it stands among every measured tool in its category and how that has moved.

Score cards are part of every paid plan.

See plans →

Each score is 0–100 on a published rubric, derived from data this site already shows. A dimension we cannot measure says so — it is never scored 0 and never enters a rank.

How this is scored
  • Shipping velocity — releases a month over the last 12 months, on a log scale where 16 a month scores 100. Needs 3 releases in the window.
  • Documentation quality — the release-notes rubric published on the analytics page, unchanged.
  • Breaking-change discipline — the share of releases carrying a breaking change (none scores 100, half or more scores 0). Only projects that label breaking changes are measured: a clean record cannot be told from an unlabelled one.
  • Maintenance risk — quiet time measured against the tool's own release cadence (quiet longer than 90 days and three typical gaps scores 0), plus the number of active maintainers and whether contributors are growing, where known. Higher is lower risk.
  • Innovation & surface — new capabilities in the feature ledger this year, API and MCP operations added, and the share of commits that are features: the mean of whichever we hold.
  • Community health (open-source tools only) — each is a direction, not a size: contributors this year against last, the bus factor now against a year ago, and issues closed per issue opened, where steady scores 50, doubled 100 and halved 0; plus the median time to a maintainer's first reply (an hour scores about 100, a month or more 0). A commercial card has no repository and carries none of the four.
  • Documented delivery (commercial tools) — when a vendor's documented new capabilities have slowed against its own baseline, that counts against the overall standing as heavily as two low dimensions. Steady delivery, or too little history to judge, leaves the standing alone; it is never a sixth score.
  • Open-source tools are scored on repository signals; commercial tools on their published changelog. Ranks compare tools within one category and one profile and need at least five measured peers; the overall standing needs three ranked dimensions. Best-in-breed or consolidator is a shape, not a quality, and is never scored.

Release-note detail over time

See how much detail GrantGuard puts in its release notes, quarter by quarter, and compare it with tracked tools in the same category. This history uses up to ten retained, reviewed notes per tool; sparse or blank older quarters may reflect that limit.

Checking release-note history…

Explore the full chart →

Releases

1 entry

  1. July 20261

    1 entry

    1. v0.1.0 Jul 7, 2026 · issue 018

      GrantGuard v0.1.0 launches: audit and clean up risky Claude Code standing permissions from the CLI.

      └──▷ GET THIS VERSION 
      $ git clone --branch v0.1.0 https://github.com/OpenVanta/GrantGuard.git
# already have the repo? check out this version:
$ git checkout v0.1.0
      └──▷ TRY IT
      Quickly audit your personal Claude Code permissions for high-risk grants — inline API keys, credential-store reads, destructive wildcards — without modifying anything.
      $ uv run grantguard.py audit
      Open the interactive browser UI to review and selectively remove flagged permissions across a specific repo's Claude settings.
      $ uv run grantguard.py ui /path/to/repo
      • New audit command scans Claude Code settings files and classifies every allow rule by risk — inline credentials, credential-store reads, destructive wildcards, unprompted remote pushes, and overbroad wildcards.
      • New ui command opens a local browser UI for interactive review and one-click removal of flagged permission grants.
      • Supports --scan for shallow discovery of .claude/settings*.json under target directories, and --deep-scan for broad recursive discovery.
      • New --fix flag on audit applies removals non-interactively, rewriting only the flagged rules while leaving all other settings intact.
      • New --tolerance flag (default or permissive) controls whether overbroad wildcard rules are flagged or treated as safe.
      +3 more show less
      • New --show-safe flag on audit prints kept (non-flagged) rules alongside flagged ones for full visibility.
      • Reads all standard Claude Code permission sources by default: ~/.claude/settings.json, ~/.claude/settings.local.json, ~/.claude.json, and the platform managed settings file.
      • Cross-platform support for macOS, Linux, and Windows with zero third-party runtime dependencies.

      8 details in Everything

    Project health

    Issue, contributor and maintainer activity in OpenVanta/GrantGuard, 2023-10 – 2026-09, as of 2026-09-25. Twelve-month figures first; the trend lines run the whole window.

    1. Contributors 4 +4 vs prior 12 months
    2. Maintainers 2 +2 joined · −0 left, last 12 months
    3. Bus factor 1 committers behind half the commits
    4. Community-filed 100% of issues opened, last 12 months
    5. Median time to close not measured issues closed, last 12 months
    6. Open backlog 1 of 1 issues ever filed
    1. Issues opened / month 0 · 0 communitymaintainers
    2. Open backlog 1
    3. Contributors / month 0
    4. Commits / month 0
    5. Stars 41
    Issue themes, last 12 months
    • other 100%

    Counts from the project's public repository and git history; no individuals are identified. Maintainers are the committers covering 80% of the last year's commits. how we measure this

    Add the badge to your README

    Maintain this project? The badge links readers straight to this page, so they can follow every release here.

    Tracked by Cyber Toolchain Choose a badge style and copy the code →
    compare.md

      Free the public facts, no sign-in Any paid plan all other sections Researcher & Business pricing history and category rank

      my-toolchain — 0 tools

        Betas, release candidates and previews. Turn it off to see final releases only. It also applies to changes you ask for over MCP when you're signed in.
        browse all tools → score card for your toolchain →
        cyber-toolchain / feedback

        Send us feedback