updated 2026-10-09 08:15 UTC

Claude Code

Repository Website by anthropics

Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands.

Coding Agents TypeScript Needs review Proprietary

Is Claude Code safe? Promising trust profile, but some evidence still deserves review.

OSSF Scorecard 6.0 / 10
Provenance None
Signed commits 68%
Last push yesterday
Advisories None known
Compare Claude Code

How does it stack up against its Coding Agents neighbours?

Pick any agent to compare →
Suggest a correction

In detail: Claude Code scores 70.4/100 (Grade B), ranked #327 of 1371 tracked open-source AI agent projects, on evidence coverage A (5 of 5 independent signal types). The public evidence: no package-provenance attestation found; OSSF Scorecard rates its supply-chain practices 6.0/10; 68% of recent commits are signed; no published advisory affects its latest release (@anthropic-ai/claude-code); last pushed 2026-10-08. Every point is earned from checkable signals — never paid placement. How scoring works →

What Claude Code can reach, and who runs it

Checked 2026-10-06, each with its source. PUBLISHER is the company's or project's own statement, which we haven't verified; CHECKED is something we observed ourselves. These facts describe the product. They don't change its score.

Who runs it

  • CHECKED Made by Anthropic. The repository's licence reserves all rights and puts use under Anthropic's Commercial Terms, so it isn't open source. github.com

What it can reach

  • PUBLISHER Interactive sessions start in Auto mode, where a classifier model approves or blocks actions. In Manual mode it starts read-only and asks before editing files or running commands. code.claude.com
  • PUBLISHER Its file tools ask before reading or writing outside the folder it started in, but a shell command you approve can write anywhere your user account can unless you turn on its sandbox. code.claude.com

Where your data goes

  • PUBLISHER Prompts and model output go to Anthropic, or to Bedrock, Google Cloud or Microsoft Foundry if you configure one. Commercial accounts: kept 30 days and not used for training. Consumer accounts: used for training when that setting is on (kept 5 years), otherwise kept 30 days. code.claude.com
  • PUBLISHER Session transcripts are stored on your machine in plain text under ~/.claude/projects/ for 30 days by default. Usage metrics exclude code, prompts and file paths and can be turned off with DISABLE_TELEMETRY=1. code.claude.com

Security practice

  • PUBLISHER Anthropic publishes its SOC 2 Type 2 report and ISO 27001 certificate in its trust center, trust.anthropic.com. Vulnerabilities go to a HackerOne program. code.claude.com
  • CHECKED Its maintainers have published 33 security advisories (26 high, 4 medium, 3 low), 23 of them in the last 12 months. The repository has a SECURITY.md. github.com
Something wrong or out of date? Suggest a correction

How Claude Code could raise its score

Each line changes one public signal and recomputes with the live scoring function. The gains don't add up exactly, because the score is capped near the top.

Already done? Suggest a correctionAdd the trust badge
Active listing, but review is needed
  • [4.1.1] no declared license (GitHub license field is null)

Chain of custody

Scanners check what the code says. This traces who ships Claude Code and whether that has changed: from the source repository, through how changes are reviewed and released, to the package you install. These are the checks for OWASP ASI04 Agentic Supply Chain Vulnerabilities.

  1. Source Partly signed

    github.com/anthropics/claude-code, last pushed 2026-10-08. 68% of the last 100 commits carry a verified signature; the rest can't be tied to a verified identity.

  2. Review and release Mixed

    OpenSSF Scorecard rates the repository's practices 6.0/10 (scanned Sep 30, 2026). The checks that decide who can get a change released:

    • Code Review6
    • Branch Protection5
    • Signed Releases8
    • Dangerous Workflow10
    • Token Permissions0
    • Pinned Dependencies3
    All 18 Scorecard checks
    Binary-Artifacts 10
    Branch-Protection 5
    CI-Tests 7
    CII-Best-Practices 0
    Code-Review 6
    Contributors 10
    Dangerous-Workflow 10
    Dependency-Update-Tool 0
    Fuzzing 0
    License 9
    Maintained 10
    Packaging -1
    Pinned-Dependencies 3
    SAST 0
    Security-Policy 10
    Signed-Releases 8
    Token-Permissions 0
    Vulnerabilities 10

  3. Published packages Linked, not attested

    Without a build attestation, nothing proves the published files were built from this repository.

  4. What you install today No known advisories

    Latest release checked: @anthropic-ai/claude-code 2.1.293. No published advisory affects it (OSV, checked 2026-10-09).

Changes to this chain

No change to package provenance, package source links, Scorecard coverage or license in HVTracker's daily snapshots of Claude Code.

Evidence behind this score

Coverage A: 5 of 5 independent evidence types found.

  • GitHub repository data
  • Package downloads
  • Supply-chain checks
  • Public actions
  • Community mentions

Verify this score yourself

Every build re-issues Claude Code’s score as an Ed25519-signed credential, valid for 7 days. You can check it offline against HVTracker’s published key; if anyone changes a number after signing, verification fails.

Signed credential (JSON) How to verify

What this score doesn’t check

It covers who ships the code, not what the code does. It doesn’t read tool descriptions or prompts for injected instructions, watch runtime behaviour, or find bugs nobody has disclosed yet. For that, run a content scanner before you connect it, such as Cisco MCP Scanner or Snyk Agent Scan.

How Claude Code compares in Coding Agents

  1. #15 DeepSeek Harness 71.8 +1.4
  2. #16 Tau 71.7 +1.3
  3. #17 Claude Code 70.4 this agent
  4. #18 mini-SWE-agent 70.1 −0.3
  5. #19 Goose 69.2 −1.2

Bars show each HVTrust score; the tick marks Claude Code’s 70.4.

Where the 70.4 comes from

HVTrust dimensions vs the Coding Agents average

70.4 / 100 · 100.0% confidence

Claude Code Coding Agents average (77 agents)

Safety / Integrity50% OSSF Scorecard · 30% provenance · 20% signed commits
10.9 / 25
2.1 above avg 8.8
Identity / Provenance60% listing status · 40% build provenance
10.8 / 18
0.9 below avg 11.7
Transparency50% declared license · 50% OSSF Scorecard
5.1 / 17
6.5 below avg 11.6
Maintenance60% last-push freshness · 40% commit activity
19.9 / 20
5.5 above avg 14.4
AdoptionLog-scaled stars · package downloads
20.0 / 20
8.1 above avg 11.9

Quick Trust Read

What Would Improve It
Publish package provenance or release attestations for stronger supply-chain evidence.
Recent Changes
2026-09-23
Rank Moved
Rank dropped 190 spots (#115 → #305)
2026-09-21
Rank Moved
Rank rose 185 spots (#300 → #115)
2026-09-17
Rank Moved
Rank dropped 15 spots (#286 → #301)
Maintainer Checklist
Declare an open license Publish a clear OSI-approved license so usage and maintenance terms are independently verifiable.
Raise Scorecard signals Current OSSF Scorecard is 6.0/10. Tighten the weakest checks to improve public safety evidence.
Publish provenance Add package provenance or release attestations so users can verify where shipped artifacts came from.
Review Flags
[4.1.1] no declared license (GitHub license field is null)
Compare with another agent Get email alerts for Claude Code View Coding Agents Report missing data
100.0
Activity sub-score · out of 100
#17
In Coding Agents

How to read this: HVTrust (0–100) weighs supply-chain signals (provenance, OSSF Scorecard, signed commits, open license) alongside real-world adoption. Grade B reflects the trust score band: A ≥ 80, B ≥ 65, C ≥ 50, D < 50. Evidence coverage A is separate — it grades how many independent signal types back the score (5 of 5), so a high score on thin evidence stays visible. Full methodology →

Signals refreshed 2026-10-09 08:15 UTC · Repo last pushed yesterday

Rank Trend

2026-09-29 2026-10-09

Activity & Reach

Stars
149.8k
Forks
25.8k
Last Push
2026-10-08
yesterday
Commits (4 wk)
159
Downloads (7d)
11,675,477
npm
HN mentions (30d)
488
Open Issues
13925
Rank Change
▼1
was #326

Analysis

Activity Inputs

100.0 / 100
StarsRepository reach
30 / 30
FreshnessLast push recency
24.9 / 25
ActivityRecent commits
25 / 25
CommunityFork signal
20 / 20

Common questions about Claude Code

Proprietary software. A lower HVTrust score reflects fewer public supply-chain artifacts (open license, build provenance, signed commits) — not a security finding. Claude Code may have internal security practices not visible through public signals.
Claude Code has a mixed signal profile. Some trust indicators are present, others are missing. Whether it is safe for your use case depends on which gaps matter to you — review the breakdown below before adopting in production.
Does Claude Code publish package provenance?
No published build provenance is currently detected for Claude Code. This is common for open-source projects but means consumers cannot independently verify that the package on the registry matches the GitHub source.
Does Claude Code have an OpenSSF Scorecard?
Claude Code has an OpenSSF Scorecard score of 6.0/10. The Scorecard checks for branch protection, signed releases, dependency updates, fuzzing, code review, and other supply-chain hygiene items. See the full check breakdown on this page.
Is Claude Code actively maintained?
Actively maintained. The repository was pushed to within the last 1 day(s).
What license does Claude Code use?
Claude Code ships under no SPDX license detected. A declared, OSI-approved license is one of the transparency signals HVTrust scores.
Are Claude Code's commits signed?
68% of the last 100 commits to Claude Code are verified-signed (GPG, SSH, S/MIME, or GitHub's signing flow). Signed commits help confirm that code was authored by who the commit claims.

Not a safety endorsement. HVTracker describes what public signals show, not whether a project is safe for your use case. Run your own security review before adopting in production.

AI agent surface

MCP, providers, tool surface
Scored in HVTrust

These runtime-trust fields — detected from public repo docs and manifests — contribute a bounded adjustment to this project's HVTrust score alongside supply-chain evidence. The exact values each field can add or subtract are documented in the methodology → Compare this surface across every listed agent in the capability matrix →

MCP Server Support
None detected
No MCP server signal detected.
Detailed evidence is not shown in the public view.
External Service Dependencies
medium confidence
1 detected
Public provider/service dependencies detected.
Credential signal: No explicit API-key/config marker detected.
Tool / Plugin Surface
high confidence
Declared
Declared plugin/integration surface detected.
Detailed evidence is not shown in the public view.
Package Provenance Drift
high confidence
Match
Published package metadata matches the tracked repo
Detailed evidence is not shown in the public view.
  • MCP signal live
  • External deps live
  • Tool / plugin surface live
  • Package provenance drift live
How this surface has changed

Detected changes to Claude Code's runtime surface and supply-chain posture, from daily public-signal snapshots. A change here means our detectors see something different — a genuinely changed capability, or better evidence of an existing one.

2026-07-17
Provider Added
Runtime surface grew — new detected provider dependency: Anthropic

Maintain Claude Code?

For maintainers

HVTrust scores Claude Code from public signals only — we never contact maintainers first. If a signal is wrong, stale, or missing (provenance you publish, a Scorecard you run, signed releases), tell us and we'll review it. Corrections are public and tracked on GitHub.

Submit a correction → Add the trust badge

Public Actions

Actions (30d)
84402
Merged (30d)
73704
Merge rate
87%
Top repos (sampled)

Action counts reflect activity on public repos where this agent left detectable fingerprints. Private repo usage is invisible. Counts are approximate (GitHub Search API caps at 1,000 results per query). Methodology described in the Provenance Profile spec.

Reputation Timeline

Signal history
Rank 26Grade 4Surface 1HVTrust 1
2026-09-23
Rank Moved
Rank dropped 190 spots (#115 → #305)
2026-09-21
Rank Moved
Rank rose 185 spots (#300 → #115)
2026-09-17
Rank Moved
Rank dropped 15 spots (#286 → #301)
2026-09-14
Rank Moved
Rank rose 61 spots (#347 → #286)
2026-09-14
HVTrust Changed
HVTrust up 3.4pts (68.7 → 72.1)
2026-09-13
Rank Moved
Rank rose 60 spots (#407 → #347)
2026-09-12
Rank Moved
Rank rose 44 spots (#451 → #407)
2026-09-11
Rank Moved
Rank dropped 10 spots (#441 → #451)
2026-09-10
Grade Changed
Trust grade C → B
2026-09-10
Rank Moved
Rank rose 14 spots (#455 → #441)
2026-09-07
Rank Moved
Rank dropped 33 spots (#420 → #453)
2026-09-05
Rank Moved
Rank dropped 71 spots (#341 → #412)
2026-09-03
Rank Moved
Rank dropped 15 spots (#323 → #338)
2026-08-31
Rank Moved
Rank dropped 21 spots (#293 → #314)
2026-08-30
Rank Moved
Rank rose 10 spots (#303 → #293)
2026-08-29
Rank Moved
Rank dropped 19 spots (#284 → #303)
2026-08-26
Rank Moved
Rank dropped 16 spots (#260 → #276)
2026-08-24
Rank Moved
Rank dropped 19 spots (#244 → #263)
2026-08-22
Rank Moved
Rank rose 16 spots (#262 → #246)
2026-08-19
Rank Moved
Rank dropped 12 spots (#247 → #259)
2026-08-17
Rank Moved
Rank rose 20 spots (#266 → #246)
2026-08-16
Rank Moved
Rank dropped 10 spots (#256 → #266)
2026-08-13
Rank Moved
Rank dropped 18 spots (#237 → #255)
2026-08-12
Rank Moved
Rank rose 15 spots (#252 → #237)
2026-08-10
Rank Moved
Rank dropped 13 spots (#242 → #255)
2026-08-08
Rank Moved
Rank dropped 40 spots (#195 → #235)
2026-08-04
Rank Moved
Rank rose 14 spots (#173 → #159)
2026-07-26
Grade Changed
Trust grade B → C
2026-07-17
Provider Added
Runtime surface grew — new detected provider dependency: Anthropic
2026-07-14
Grade Changed
Trust grade C → B
2026-07-13
Grade Changed
Trust grade B → C
2026-07-13
Rank Moved
Rank dropped 17 spots (#120 → #137)
GitHub → Website → npm: @anthropic-ai/claude-code

Embed Badge Badge guide for maintainers →

For maintainers
HVTrust 70.4 Grade B
Markdown:
[![HVTrust](https://hvtracker.net/badge/claude-code.svg)](https://hvtracker.net/agents/claude-code)
HTML:
<a href="https://hvtracker.net/agents/claude-code"><img src="https://hvtracker.net/badge/claude-code.svg" alt="HVTrust"></a>

Other agents in Coding Agents

Data sources
GitHub REST API (repo, commits, stars, forks, license) · npm Registry (downloads, provenance) · OpenSSF Scorecard CLI · Algolia HN Search API · GitHub Search API (fingerprint actions)
Each agent's signals refresh once daily across 6 staggered batches. Methodology v4.4 · Raw JSON