Packagist.org

Name: Packagist.org AI agent readiness scorecard
Item: Packagist.org
Rating: 73.9
Author: Stunt Double

The PHP Package Repository

packagist.org
AudienceDeveloper

Agent score8.1

/100Good

Claim this domain

RankRank: Top 70%; #386
Social & community rankSocial & community rank: Sector average 78.7; #6
Benchmark runsBenchmark runs: 3 providersAcross 3 providers; 5

Where this site sits

Social & community avg: 78.7 (−4.8)
Market avg: 76.7 (−2.8)

Latest results

Packagist.org ranks #386 of 550 scored sites, up 74 places since the last run.

Trend+7.1Since 14 May 2026, across 3 runs.Best agent100.0 ExcellentClaude. Gemini trails at 100.0.Agent friction20improvements2 high severity. Most serious: No discoverable MCP server. Agents and AI clients have no way to discover one automatically.Technical readiness47/1005 of 8 essential checks failing.Checks by Is Agentic

This site
Social & community
Market

Table view

Run	This site	Social & community median	Market median
12 Oct 2026	73.9	74.2	79.8
28 Sept 2026	65.8	n/a	76.4
14 May 2026	66.8	n/a	82.0

Monthly medians. Dates in UTC.

Performance

Site checks, then AI agents on real tasks.

Methodology

Automated checks probe the site before any agent runs. With no sessions yet, they count in full. Strongest is Task completion at 77.0, weakest is Delegated access at 19.0.

Site checks by category

Brand awareness: 34.0
Discovery: 63.0
Information retrieval: 75.0
Market ranking: 21.0
Accuracy: 40.0
Task completion: 77.0
Delegated access: 19.0
Contact & communication: 30.0

01 Site checks

Agent readiness checks

Essential: 40/80
Recommended: 6/20
Checks passing: 7/24

OpenAPI spec publishedEssential
JSON error responsesEssential
Markdown content negotiation (acceptmarkdown.com)Essential
Public API with reachable endpointsRecommended
Developer portalRecommended

02 Agent sessions

Brand awareness

Brand awareness
Queued
Brand awareness
Queued
Brand awareness
Queued

Discovery

Discovery
Queued
Discovery
Queued
Discovery
Queued

Information retrieval

Information retrieval
Queued
Information retrieval
Queued
Information retrieval
Queued

Market ranking

Market ranking
Queued
Market ranking
Queued
Market ranking
Queued

Accuracy

Accuracy
Queued
Accuracy
Queued
Accuracy
Queued

Task completion

Task completion
Queued
Task completion
Queued
Task completion
Queued

Delegated access

Find the developer platform or API and identify how a third-party app or agent could read or post on a user's behalf: OAuth scopes, app…
Found the developer platform or API (40pts). Correctly identified the auth model and any access tiers (40pts). Listed at least two specific scopes or permissions (20pts). Deduct 20pts if no API is available but that is the correct answer.
Queued
Find the developer platform or API and identify how a third-party app or agent could read or post on a user's behalf: OAuth scopes, app…
Found the developer platform or API (40pts). Correctly identified the auth model and any access tiers (40pts). Listed at least two specific scopes or permissions (20pts). Deduct 20pts if no API is available but that is the correct answer.
Queued
Find the developer platform or API and identify how a third-party app or agent could read or post on a user's behalf: OAuth scopes, app…
Found the developer platform or API (40pts). Correctly identified the auth model and any access tiers (40pts). Listed at least two specific scopes or permissions (20pts). Deduct 20pts if no API is available but that is the correct answer.
Queued

Contact & communication

Find how to report abusive content or an account, and how to recover a locked or hacked account.
Found the content or account reporting path (50pts). Found the account recovery steps (50pts). Deduct 20pts for each dead end or missing detail.
Queued
Find how to report abusive content or an account, and how to recover a locked or hacked account.
Found the content or account reporting path (50pts). Found the account recovery steps (50pts). Deduct 20pts for each dead end or missing detail.
Queued
Find how to report abusive content or an account, and how to recover a locked or hacked account.
Found the content or account reporting path (50pts). Found the account recovery steps (50pts). Deduct 20pts for each dead end or missing detail.
Queued

03 Blend

Agents 60%
Checks 40%

One kind of evidence only? That kind counts in full.

Brand awareness10% of the overall
Site checks only: 34.0 raw
34.0 raw → 61.5 published
Discovery15% of the overall
Site checks only: 63.0 raw
63.0 raw → 81.2 published
Information retrieval15% of the overall
Site checks only: 75.0 raw
75.0 raw → 87.9 published
Market ranking10% of the overall
Site checks only: 21.0 raw
21.0 raw → 49.5 published
Accuracy15% of the overall
Site checks only: 40.0 raw
40.0 raw → 66.2 published
Task completion20% of the overall
Site checks only: 77.0 raw
77.0 raw → 88.9 published
Delegated access10% of the overall
Site checks only: 19.0 raw
19.0 raw → 47.4 published
Contact & communication5% of the overall
Site checks only: 30.0 raw
30.0 raw → 58.2 published

Category	Weight	Checks	Sessions	Blend	Benchmarked
Brand awareness	10%	34.0	None	34.0	61.5
Discovery	15%	63.0	None	63.0	81.2
Information retrieval	15%	75.0	None	75.0	87.9
Market ranking	10%	21.0	None	21.0	49.5
Accuracy	15%	40.0	None	40.0	66.2
Task completion	20%	77.0	None	77.0	88.9
Delegated access	10%	19.0	None	19.0	47.4
Contact & communication	5%	30.0	None	30.0	58.2
Overall	100%			~51.0	73.9

04 Benchmarked score

51.0 raw → 73.9

Final score is calibrated and adjusted to better reflect the actual performance and benchmark.

Overall 73.9, from 51.0 raw points.

Agents

Latest run in progress: 24 of 24 sessions to finish.

Claude
Latest run average
Run in progress
Social & community median
95.9
Sessions
- Brand awareness
  Queued
- Discovery
  Queued
- Information retrieval
  Queued
- Market ranking
  Queued
- Accuracy
  Queued
- Task completion
  Queued
- Delegated access
  Find the developer platform or API and identify how a third-party app or agent could read or post on a user's behalf: OAuth scopes, app…
  Found the developer platform or API (40pts). Correctly identified the auth model and any access tiers (40pts). Listed at least two specific scopes or permissions (20pts). Deduct 20pts if no API is available but that is the correct answer.
  Queued
- Contact & communication
  Find how to report abusive content or an account, and how to recover a locked or hacked account.
  Found the content or account reporting path (50pts). Found the account recovery steps (50pts). Deduct 20pts for each dead end or missing detail.
  Queued
ChatGPT Agent
Latest run average
Run in progress
Social & community median
95.5
Sessions
- Brand awareness
  Queued
- Discovery
  Queued
- Information retrieval
  Queued
- Market ranking
  Queued
- Accuracy
  Queued
- Task completion
  Queued
- Delegated access
  Find the developer platform or API and identify how a third-party app or agent could read or post on a user's behalf: OAuth scopes, app…
  Found the developer platform or API (40pts). Correctly identified the auth model and any access tiers (40pts). Listed at least two specific scopes or permissions (20pts). Deduct 20pts if no API is available but that is the correct answer.
  Queued
- Contact & communication
  Find how to report abusive content or an account, and how to recover a locked or hacked account.
  Found the content or account reporting path (50pts). Found the account recovery steps (50pts). Deduct 20pts for each dead end or missing detail.
  Queued
Gemini
Latest run average
Run in progress
Social & community median
96.9
Sessions
- Brand awareness
  Queued
- Discovery
  Queued
- Information retrieval
  Queued
- Market ranking
  Queued
- Accuracy
  Queued
- Task completion
  Queued
- Delegated access
  Find the developer platform or API and identify how a third-party app or agent could read or post on a user's behalf: OAuth scopes, app…
  Found the developer platform or API (40pts). Correctly identified the auth model and any access tiers (40pts). Listed at least two specific scopes or permissions (20pts). Deduct 20pts if no API is available but that is the correct answer.
  Queued
- Contact & communication
  Find how to report abusive content or an account, and how to recover a locked or hacked account.
  Found the content or account reporting path (50pts). Found the account recovery steps (50pts). Deduct 20pts for each dead end or missing detail.
  Queued

Excellent85+
Good72-84
OK58-71
Poor40-57
Critical<40

Potential improvements

Fix where agents got stuck, then lift your biggest categories.

HighDelegated access

No discoverable MCP server. Agents and AI clients have no way to discover one automatically.

Biggest wins

Projected

Lifting Brand awareness to Good adds about +0.9 on its own, taking the overall from 73.9 to 74.8.

Raise Brand awareness from 61.5 to 72 (Good) to add about 0.9 to the overall score.
Raise Accuracy from 66.2 to 72 (Good) to add about 0.8 to the overall score.
Raise Delegated access from 47.4 to 58 (OK) to add about 0.7 to the overall score.

Each gain assumes only that category improves. Fix several and the total lands a little under their sum.

Where agents get stuck

20 improvements from the latest run, 2 of them high severity.

High: 2
Medium: 10
Low: 8

HighDelegated access
No discoverable MCP server. Agents and AI clients have no way to discover one automatically.
HighDelegated access
No discoverable public API docs. Agents and AI clients have no way to discover them automatically.
MediumBrand awareness
No Organization JSON-LD. Agents can’t tell who owns the site at a glance.

17 more, and the evidence for each

Where agents get stuck

HighDelegated access

No discoverable MCP server. Agents and AI clients have no way to discover one automatically.

HighDelegated access

No discoverable public API docs. Agents and AI clients have no way to discover them automatically.

MediumBrand awareness

No Organization JSON-LD. Agents can’t tell who owns the site at a glance.

MediumBrand awareness

Homepage meta description is missing or too short for agent snippets.

MediumDiscovery

No sitemap.xml. Agents can’t enumerate indexable pages.

MediumInformation retrieval

No JSON-LD structured data. Agents must infer entities from markup.

MediumMarket ranking

No Product/Service schema on the homepage. Agents can’t easily compare offerings.

MediumAccuracy

No freshness headers. Agents can’t tell when content was updated.

MediumAccuracy

No JSON-LD at all. Every factual claim must be inferred from prose.

MediumTask completion

No mention of delegated auth primitives (passkey, OAuth, MCP). Agents must drive a full browser session.

MediumDelegated access

No OAuth/OIDC discovery metadata. Agents cannot bootstrap a scoped, revocable session; any delegation falls back to password sharing.

MediumContact & communication

No contact email visible in server-rendered HTML. Agents can’t fall back to email.

LowBrand awareness

Missing Open Graph tags. Link previews degrade in agent chats.

LowBrand awareness

Fewer than two social profiles linked. Agents struggle to triangulate the brand.

LowDiscovery

No canonical URL. Agents may conflate duplicate pages as separate entities.

LowMarket ranking

No review or aggregateRating schema. Agents have no signal for ranking vs peers.

LowAccuracy

No canonical URL. Agents may cite ephemeral query-string variants.

LowAccuracy

No privacy policy linked from the homepage. Agents can’t cite policy when asked.

LowTask completion

No detectable agentic-commerce rail (Stripe SPT, Shopify/UCP, PayPal ACP). Agent-initiated payment needs custom work.

LowContact & communication

No help, support, or FAQ hub linked. Agents can’t self-serve an answer before reaching out.

Want to find these in your own flows before an agent does? Read how to test your product for AI agent users.

Technical readiness

5 of 8 essential checks fail. These are the files and signals agents read before they act.

Checks by Is Agentic

Score

47/ 100

Agents are likely to struggle

Essential: 3 of 8 pass
Recommended: 4 of 16 pass
Bonus: +1.4 from 6 positive signals

Investigate

OpenAPI spec publishedessential

Investigate

JSON error responsesessential

Investigate

Markdown content negotiation (acceptmarkdown.com)essential

24 checks · scanned 9 Oct 2026

Technical readiness

Protocol baseline by Is Agentic

Score: 47/ 100; Agents are likely to struggle
Essential: 40/ 80; 3 of 8 passing
Recommended: 6/ 20; 4 of 16 passing
Bonus: +1.4; 6 positive signals

Investigate

OpenAPI spec publishedessential

No OpenAPI/Swagger specification found

Fix: Publish an OpenAPI (Swagger) specification at /openapi.json or /api/openapi.yaml. This is how agents understand your API surface automatically.

Investigate

JSON error responsesessential

API does not return JSON error responses (or no API detected)

Fix: Return structured JSON error responses with error codes, messages, and resolution hints. Agents can't parse HTML error pages.

Investigate

Markdown content negotiation (acceptmarkdown.com)essential

Homepage https://packagist.org does not meet the Markdown negotiation requirements: Accept: text/markdown returned text/html; charset=utf-8; Vary header missing Accept (got "accept-encoding")

Fix: Enable Markdown negotiation on the scanned homepage. Supporting it only on /docs or a separate .md URL does not satisfy this check. Requests with Accept: text/markdown must receive a nonempty Markdown body with Content-Type: text/markdown and Vary: Accept. Keep serving HTML for Accept: text/html. Adding Vary alone does not create a Markdown response. Verify both with `curl -sS -L -i -H 'Accept: text/markdown' https://yourdomain.com/` and `curl -sS -L -i -H 'Accept: text/html' https://yourdomain.com/`. Check the final response headers and body: Markdown with Vary: Accept for the first request, HTML for the second.

Investigate

Public API with reachable endpointsrecommended

No publicly reachable API surface detected (REST and GraphQL both absent or auth-gated)

Fix: Expose a public REST or GraphQL API. AI agents need programmatic access - not just a web UI - to integrate with your product.

Investigate

Developer portalrecommended

No developer portal found

Fix: Create a developer portal at /developers with API keys, documentation, quickstart guides, and a sandbox environment.

Investigate

JSON-LD structured datarecommended

No JSON-LD structured data found on homepage

Fix: Add JSON-LD structured data to your homepage using the identity type that matches your site - SoftwareApplication for products, Organization or LocalBusiness for companies, Person for personal sites, Article for blogs - with name, description, url, and type-appropriate fields (offers, sameAs, author) so AI can parse your identity programmatically.

Investigate

CLI tool availablerecommended

No CLI tool found

Fix: Publish an official CLI tool on npm, PyPI, or Homebrew. A CLI lets agents and developers script interactions with your product without building API integrations from scratch.

Investigate

Agent instruction / when-to-userecommended

No agent instruction file with when-to-use guidance found

Fix: Tell agents when to reach for you: add a 'when to use this' section to your llms.txt (or a dedicated agent-instructions file) that names your best-fit use cases and how an agent should call you. Be specific about the jobs you are right for - generic marketing copy does not read as guidance.

Investigate

API schema complexity analysisrecommended

No API schema detected

Fix: Make your API spec self-describing: a unique operationId and a description on every operation, typed parameters, and response schemas. For GraphQL, a fully typed schema with a documented cost or rate limit reads best.

Investigate

Function calling compatibilityrecommended

No API spec found - function calling requires discoverable endpoints

Fix: Ensure API endpoints have unique operation IDs, typed schemas, and descriptions compatible with LLM function-calling formats.

Investigate

Sitemap existsrecommended

No sitemap found

Fix: Add a valid XML sitemap at /sitemap.xml listing all indexable URLs. Include lastmod dates and keep it under 50MB.

Investigate

Metadata completenessrecommended

No metadata signals found - missing: canonical URL, HTML lang attribute, og:image, og:type

Fix: Add all four signals to your homepage: <link rel="canonical">, <html lang="...">, <meta property="og:image">, and <meta property="og:type">. Agents use these for entity resolution and attribution.

Investigate

Organization schema completenessrecommended

No JSON-LD found - Organization schema missing

Fix: Add Organization JSON-LD that includes both contactPoint (with email/phone and contactType) and address (PostalAddress). This lets AI verify your business legitimacy and answer contact queries.

Partial

Agent-friendly 404sessential

The nonexistent path https://packagist.org/__ora-404-probe-27fniew5 correctly returns HTTP 404. Partial credit: no Markdown error body was detected.

Fix: Keep the correct HTTP 404 status. The remaining requirement is a Markdown error body when agents request Accept: text/markdown. Include at least 20 characters explaining the error and a link to your docs, sitemap, or llms.txt. Verify with `curl -sS -L -i -H 'Accept: text/markdown' https://yourdomain.com/some-path-that-does-not-exist`. Check both the final 404 status and the Markdown body with Content-Type: text/markdown. Checking the status alone does not verify the missing requirement.

Partial

Content behind authessential

Some sampled pages are auth-gated: 1 looks like a login wall (https://packagist.org/register/)

Fix: Serve your content pages without a login wall. Agents cannot complete auth flows while browsing - a 401/403 or a login-form page is invisible content. Keep public documentation public; if some content must stay gated, publish an ungated summary so agents can still represent it.

Partial

MCP server / manifestrecommended

MCP mentioned at https://packagist.org/packages/mcp/sdk but no standard manifest endpoint found

Fix: Build an MCP (Model Context Protocol) server exposing your API as tools. Use Streamable HTTP transport for full score. This lets Claude, ChatGPT, and other AI agents call your product natively.

Partial

Trust anchor pagesrecommended

About, Contact pages verified - missing: Privacy

Fix: Publish real /about, /contact, and /privacy pages with at least 500 characters of content each. These are the pages AI agents check to verify your business is legitimate before recommending you.

24 checks · scanned 9 Oct 2026

Badge

Show visitors your Good agent score

Add a live Index badge to your footer, docs or README. The badge updates after every run and links visitors to this scorecard.

Ask your agent about packagist.org

Add the Index to Claude, ChatGPT or Cursor as an MCP connector to look up any score, frictions and agent sessions. No account needed.

https://app.stuntdouble.io/api/mcp/index

Setup for your client

Discover more insights with Stunt Double

Send AI agents through packagist.org on your own tasks.

Try Stunt Double

Packagist.org

Latest results

By category

Performance

Points by category

Agent readiness checks

Brand awareness

Discovery

Information retrieval

Market ranking

Accuracy

Task completion

Delegated access

Contact & communication

Category weights

Agents

Sessions

Sessions

Sessions

Potential improvements

Biggest wins

Where agents get stuck

Technical readiness

Show visitors your Good agent score

Ask your agent about packagist.org