Not in the top 1,000 by GitHub stars
agentrhq/authsome
Credential gateway for AI agents. Log in once via Oauth2 or API Key. Every agent stays authenticated — headless, no SaaS, agents never see your credentials.
History
Review weekly changes to SKILL.md before updating. Dates mark the weeks in which we observed each version; checks apply to the recorded commits.
View weekly changes
skills/authsome/SKILL.md
Week of Nov 2, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 0d756011a6fe
No earlier version was observed for comparison.
View on GitHubAdd to my skills
Verified commit 0d756011a6fecebd28b8a7e0d91d9a3d711f1991
git clone --filter=blob:none --no-checkout https://github.com/agentrhq/authsome.git .hotskills/agentrhq-authsome && git -C .hotskills/agentrhq-authsome checkout --detach 0d756011a6fecebd28b8a7e0d91d9a3d711f1991
# Copy each selected skill directory from .hotskills/agentrhq-authsome into .claude/skills/<skill-name>/git clone --filter=blob:none --no-checkout https://github.com/agentrhq/authsome.git .hotskills/agentrhq-authsome && git -C .hotskills/agentrhq-authsome checkout --detach 0d756011a6fecebd28b8a7e0d91d9a3d711f1991
# Copy each selected skill directory from .hotskills/agentrhq-authsome into .agents/skills/<skill-name>/104GitHub stars
4stars in 7 days
1skill in this repository
0commits in 30 days
Heat
3.2- Star growth, last 7 daysNot available
- Star growth, last 24 hoursNot available
- Rank movement, last 7 daysNot available
- Commits, last 30 daysNot available
- Accounts asking about it through MCP, last 7 daysNot available
- Hot votes this weekNot available
Heat never changes the order by stars, and sponsorship never touches it. How heat works
Stars over time
30-day change not yet available: not enough history.
Individual skills in this repository
- authsome
skills/authsome
How it works
authsome
When your agent uses it
Use this to access external services/CLIs/APIs: Gmail/gh/Github/Stripe etc. or when running any bash command, script, or curl/wget that makes outbound HTTP calls. Make HTTP requests directly and the gateway injects credentials automatically.
What it loads into context
Frontmatter
70 tokensAlways available to the agent
SKILL.md
1,086 tokensWhen the skill is activated
Reference documents
1,000 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- How to Access External Services
- Making Requests
- Login Flow
- When a Request Fails
- Adding a new provider
- Troubleshooting & Help
- Rules
Scripts executed separately
evals/generate_report.py
Files
../../.claude-plugin/marketplace.jsonasset836 bytesSKILL.mdcore4,698 bytesSKILL.mdcore4,698 bytesSKILL.mdcore4,698 bytesSKILL.mdcore4,698 bytesSKILL.mdcore4,698 bytesSKILL.mdcore4,698 bytesSKILL.mdcore4,698 bytesevals/evals.jsonasset6,474 bytesevals/generate_report.pyruns9,180 bytesreferences/adding-provider.mddoc1,498 bytesreferences/feedback.mddoc2,757 bytes
Read SKILL.md
# Authsome
Your outbound HTTPS traffic is transparently proxied through the Authsome,
a proxy service that injects stored credentials as HTTP headers. You never
see or handle credential values directly.
## How to Access External Services
You have direct HTTP access to external APIs. OAuth apps (Gmail, GitHub,
Google Calendar, Google Drive, etc.) and API key services are all available
through the gateway. Just make the request by appending `authsome run -- ` to
any bash command; the gateway injects credentials if the app is connected.
## Making RequestsSkills included
- authsome
Repository facts
- License
- MIT
- Created
- Apr 16, 2026
- Last pushed
- Jul 24, 2026
- Forks
- 14
- Open issues
- 8
- Project activity
- Slowing
Before you install
Network or broad shellAutomated signals, not a guarantee
Checked Nov 6, 2026
Not fully analyzed: some files were too long, unreadable or missing. This never lowers the level.
Show 6 notes
skills/authsome/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/authsome/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/authsome/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/authsome/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/authsome/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/authsome/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed
- No broad tool permissions
- No dangerous shell commands
- 33 network referencesin 6 files: .claude-plugin/marketplace.json:3 authsome/evals/evals.json:30 authsome/references/feedback.md:51 and 3 more
- No hidden or obfuscated text
- No instruction overrides
- 1 executable scriptin 1 file: authsome/evals/generate_report.py:1
- 1 reference to secretsin 1 file: authsome/evals/evals.json:52
- No requests to vote in Hot
Show all 35 signals by file
.claude-plugin/marketplace.json
- Network accessline 3
"owner": { "name": "Agentr", "url": "https://agentr.dev" - Network accessline 4
"name": "Agentr", "url": "https://agentr.dev" }, - Network accessline 5
"url": "https://agentr.dev" - Network accessline 17
"author": { "name": "Agentr", "url": "https://github.com/agentrhq" - Network accessline 18
"name": "Agentr", "url": "https://github.com/agentrhq" }, - Network accessline 19
"url": "https://github.com/agentrhq" - Network accessline 20
}, "homepage": "https://authsome.ai", "license": "MIT", - Network accessline 21
"homepage": "https://authsome.ai",
skills/authsome/SKILL.md
- Network accessline 3
version: 0.2.0 description: | Use this to access external services/CLIs/APIs: Gmail/gh/Github/Stripe etc. or when running any bash command, script, or curl/wget that makes outbound HTTP calls. Make HTTP requests directly and the gateway i - Network accessline 4
description: | Use this to access external services/CLIs/APIs: Gmail/gh/Github/Stripe etc. or when running any bash command, script, or curl/wget that makes outbound HTTP calls. Make HTTP requests directly and the gateway injects credenti - Network accessline 5
Use this to access external services/CLIs/APIs: Gmail/gh/Github/Stripe etc. or when running any bash command, script, or curl/wget that makes outbound HTTP calls. Make HTTP requests directly and the gateway injects credentials automatically - Network accessline 27
authsome run -- <command> authsome run -- curl -s "https://api.github.com/user/repos?per_page=10" - Network accessline 28
authsome run -- curl -s "https://api.github.com/user/repos?per_page=10"
skills/authsome/evals/evals.json
- Network accessline 30
"environment": "github is connected via authsome (authsome list shows github as connected — state carried over from test 2).", "outcome": "The agent successfully fetches and displays the user's GitHub repositories using authsome (not - Network accessline 31
"outcome": "The agent successfully fetches and displays the user's GitHub repositories using authsome (not via gh CLI or direct API calls)", "trajectory_efficiency": "2 meaningful steps: (1) calls the GitHub repos API directly via aut - Network accessline 32
"trajectory_efficiency": "2 meaningful steps: (1) calls the GitHub repos API directly via authsome run -- curl https://api.github.com/user/repos without first running authsome list or authsome login. (2) to format the output", - Network accessline 48
"name": "Scenario 3 - API Key (firecrawl not configured)", "optional": true, "prompt": "Scrape the content of https://firecrawl.dev using Firecrawl", - Network accessline 49
"optional": true, "prompt": "Scrape the content of https://firecrawl.dev using Firecrawl", "agent": "claude", - Network accessline 50
"prompt": "Scrape the content of https://firecrawl.dev using Firecrawl", - Access to secretsline 52
[Credential-related source omitted]
skills/authsome/evals/generate_report.py
- Executable scriptline 1
#!/usr/bin/env python3 - Network accessline 30
font_url = "https://fonts.googleapis.com/css2?family=Poppins:wght@500;600&family=Lora:wght@400;500&display=swap" - Network accessline 37
<meta name="viewport" content="width=device-width, initial-scale=1.0"> <title>{html.escape(skill_name)} — Eval Report</title> <link rel="preconnect" href="https://fonts.googleapis.com"> - Network accessline 38
<title>{html.escape(skill_name)} — Eval Report</title> <link rel="preconnect" href="https://fonts.googleapis.com"> <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin> - Network accessline 39
<link rel="preconnect" href="https://fonts.googleapis.com"> - Network accessline 40
<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
skills/authsome/references/adding-provider.md
- Network accessline 10
3. **Guide the user based on the chosen method:** - **API key** — tell them the steps to obtain the API key from the provider's developer console. - **OAuth app** — guide the user to create an OAuth app (client ID + secret) in the provider' - Network accessline 12
- **OAuth app** — guide the user to create an OAuth app (client ID + secret) in the provider's developer console, and share the redirect URL (`http://127.0.0.1:7998/auth/callback/oauth`). - Network accessline 16
4. **Write and register the provider JSON** — follow the [provider registration guide](https://raw.githubusercontent.com/agentrhq/authsome/main/docs/register-provider.md) to write the provider JSON. Save the file to a local path (e.g. `/tmp
skills/authsome/references/feedback.md
- Network accessline 51
Proposed Solution: Add a note to the Making Requests section: If NO_PROXY contains 127.0.0.1, use: authsome run -- curl --proxy "$HTTPS_PROXY" --proxy-cacert "$SSL_CERT_FILE" \ - Network accessline 52
If NO_PROXY contains 127.0.0.1, use: authsome run -- curl --proxy "$HTTPS_PROXY" --proxy-cacert "$SSL_CERT_FILE" \ --cacert "$SSL_CERT_FILE" https://api.example.com/... - Network accessline 53
authsome run -- curl --proxy "$HTTPS_PROXY" --proxy-cacert "$SSL_CERT_FILE" \ - Network accessline 54
--cacert "$SSL_CERT_FILE" https://api.example.com/... - Network accessline 84
Scrub secrets (`ghp_...`, `sk-...`, API keys) before posting. If `gh` is unavailable, report at https://github.com/agentrhq/authsome/issues. - Network accessline 85
If `gh` is unavailable, report at https://github.com/agentrhq/authsome/issues.
How the signals fit together
Facts from a static read of the files at the pinned commit. They describe what the code does, not why.
Some files could not be read; the facts below still hold.
Declared and executed: 2 facts
- Contacts fonts.googleapis.com, which SKILL.md, the README and the plugin manifest do not mentionauthsome/evals/generate_report.py:30High confidence
- Contacts fonts.gstatic.com, which SKILL.md, the README and the plugin manifest do not mentionauthsome/evals/generate_report.py:40High confidence
Signal confidence: 20 facts
- Network access in the frontmatter of the SKILL.mdauthsome/SKILL.md:3Low confidence
- Network access in the frontmatter of the SKILL.mdauthsome/SKILL.md:4Low confidence
- Network access in the frontmatter of the SKILL.mdauthsome/SKILL.md:5Low confidence
- Network access in code of the SKILL.mdauthsome/SKILL.md:27High confidence
- Network access in code of the SKILL.mdauthsome/SKILL.md:28High confidence
- Executable script in code of a file that runsauthsome/evals/generate_report.py:1High confidence
- Network access in code of a file that runsauthsome/evals/generate_report.py:30High confidence
- Network access in code of a file that runsauthsome/evals/generate_report.py:37High confidence
- Network access in code of a file that runsauthsome/evals/generate_report.py:38High confidence
- Network access in code of a file that runsauthsome/evals/generate_report.py:39High confidence
- Network access in code of a file that runsauthsome/evals/generate_report.py:40High confidence
- Network access in prose of documentationauthsome/references/adding-provider.md:10Low confidence
- Network access in prose of documentationauthsome/references/adding-provider.md:12Low confidence
- Network access in prose of documentationauthsome/references/adding-provider.md:16Low confidence
- Network access in code of documentationauthsome/references/feedback.md:51Low confidence
- Network access in code of documentationauthsome/references/feedback.md:52Low confidence
- Network access in code of documentationauthsome/references/feedback.md:53Low confidence
- Network access in code of documentationauthsome/references/feedback.md:54Low confidence
- Network access in prose of documentationauthsome/references/feedback.md:84Low confidence
- Network access in prose of documentationauthsome/references/feedback.md:85Low confidence
Works with
| Agent | Status | Evidence | Install to |
|---|---|---|---|
| Claude Code | Native | Manifest .claude-plugin/marketplace.json | .claude/skills/<skill-name>/ |
| Codex | Native | Manifest AGENTS.md | .agents/skills/<skill-name>/ |
| Cursor | Compatible | Standard SKILL.md skills/authsome/SKILL.md | .cursor/skills/<skill-name>/ |
| Gemini CLI | Compatible | Standard SKILL.md skills/authsome/SKILL.md | .gemini/skills/<skill-name>/ |
| GitHub Copilot | Compatible | Standard SKILL.md skills/authsome/SKILL.md | .github/skills/<skill-name>/ |
| OpenCode | Compatible | Standard SKILL.md skills/authsome/SKILL.md | .opencode/skills/<skill-name>/ |
- Native The repository ships a manifest, folder or frontmatter field for this agent.
- Compatible A standard SKILL.md that this agent can read.
- Unknown No evidence either way.
Derived automatically from paths, manifests and frontmatter. Not tested on each agent.
