# pg.ddx.io: the PostgreSQL community's memory (remote · pg.ddx.io)

Every PostgreSQL mailing list since 1991, linked to the commits, code, commitfest and docs.

- Trust score: 68/100 (medium)
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-10-24

## Components

- remote · `pg.ddx.io`: 73/100, [markdown](https://verifymcp.io/servers/io-ddx-pg-archive/pg.md), [page](https://verifymcp.io/servers/io-ddx-pg-archive/pg)
- remote · `pg.ddx.io`: 68/100 (this document), [markdown](https://verifymcp.io/servers/io-ddx-pg-archive/pg-2.md), [page](https://verifymcp.io/servers/io-ddx-pg-archive/pg-2)

## Channel facts

- Endpoint: `https://pg.ddx.io/mcp/sse`
- Transports: `sse`
- Auth: `none`
- Version: `2026.10.22`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-10-24.

- **Endpoint Security**: 80/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - No authorisation is required to call this server. Every tool declares its destructiveHint and none is destructive, so open access doesn't expose one.
  - HTTPS is enforced; there's no plaintext access path.
  - The HSTS (Strict-Transport-Security) header is present.
  - DNSSEC check failed: this domain isn't protected by DNSSEC.
- **Transport & Reachability**: 40/100
  - Verified sse transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 73/100
  - 64% of prompts and resources have a non-trivial description (not blank, and not just the item's name).
  - AI-judged instruction clarity (excellent).
  - Tool/resource definitions use about 11336 tokens (~81/item across 139 items; 29 tools + 110 resources), lean.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 3/100
  - Stability observed for 1 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 100/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 100% of tool parameters carry a description.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - We read all 29 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 31 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### How do I install the pg.ddx.io: the PostgreSQL community's memory MCP server?

pg.ddx.io: the PostgreSQL community's memory is a hosted endpoint at https://pg.ddx.io/mcp/sse, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add --transport http io-ddx-pg-archive 'https://pg.ddx.io/mcp/sse'
```

### Cursor

```json
{
  "mcpServers": {
    "io-ddx-pg-archive": {
      "url": "https://pg.ddx.io/mcp/sse"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "io-ddx-pg-archive": {
      "type": "http",
      "url": "https://pg.ddx.io/mcp/sse"
    }
  }
}
```

### Codex

```toml
[mcp_servers.io-ddx-pg-archive]
url = "https://pg.ddx.io/mcp/sse"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "io-ddx-pg-archive": {
      "type": "remote",
      "url": "https://pg.ddx.io/mcp/sse",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add io-ddx-pg-archive --url 'https://pg.ddx.io/mcp/sse' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  io-ddx-pg-archive:
    url: "https://pg.ddx.io/mcp/sse"
```

### Netclaw

```json
{
  "McpServers": {
    "io-ddx-pg-archive": {
      "Transport": "http",
      "Url": "https://pg.ddx.io/mcp/sse"
    }
  }
}
```

### Vellum

```bash
assistant mcp add io-ddx-pg-archive -t streamable-http -u 'https://pg.ddx.io/mcp/sse'
```

### Other

```json
{
  "mcpServers": {
    "io-ddx-pg-archive": {
      "type": "http",
      "url": "https://pg.ddx.io/mcp/sse"
    }
  }
}
```

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-10-24 (score 68, +1)

- [functional improvement] Stability: unverified → 0.03

### 2026-10-23 (score 67)

First indexed and scored.

## MCP tools (29)

### `retrieve_context` (~241 tokens)

START HERE for a question about PostgreSQL. Searches every mailing list (inbox narrows it) with hybrid keyword + semantic ranking and returns cited passages that fit token_budget (default 4000, max 32000): one passage per thread, each the author's own words (quotes, signature and patch removed, cut at ~1200 chars) with url, inbox, message_id, from and date. found=false with a reason when the archive does not appear to discuss the question, instead of weak matches. Retrieval only: write the answer yourself. Cite each claim with its passage url or https://postgr.es/m/<message_id>; open a thread with get_thread.

Input parameters:

- `inbox` (string): Lists to search: omit (or "all") for every list, the default; one name such as pgsql-bugs; or several, comma-separated ("pgsql-hackers,pgsql-bugs"). Names from list_inboxes. Aliases: list, inbox_name.
- `question` (string, required): The question, in plain language.
- `token_budget` (number): Maximum tokens of evidence to return (default 4000, max 32000).

### `commit_buildfarm_status` (~141 tokens)

Buildfarm results for one PostgreSQL commit: every animal that built it, with branch, status (passed/failed/flapping), the failing stage and the log url, plus a {passed, failed, flapping, unknown} rollup. commit_sha is a SHA (>= 7 hex) or a branch, tag or major version (17, 9.6) for the commit it points at. Cite the commit SHA and the log urls.

Input parameters:

- `commit_sha` (string, required): A commit SHA (>= 7 hex), or a branch, tag or major version (17, 9.6) for the commit it points at

### `commit_history` (~174 tokens)

Everything linked to one commit: subject, author and date with its sha, the mail thread its Discussion: link names (participants resolved to people and organizations), the commitfest entry if tracked, and Discord mentions. commit_id takes a SHA (>= 7 hex), branch, tag, major version (17, 9.6), <ref>~N or a commit_id from another tool; repository defaults to postgres. The diff itself is not included (git_diff, or git_show_file for one file). Cite the sha and the thread's https://postgr.es/m/<message_id>.

Input parameters:

- `commit_id` (string, required): The commit: a SHA (>= 7 hex), branch, tag, major version (17, 9.6), <ref>~N, or a commit_id from another tool.

### `find_failing_patches` (~174 tokens)

Commitfest entries whose latest cfbot build fails on at least one platform: cf_id, title, entry_status and commitfest_number. entry_status filters (default the active reviewable statuses, e.g. needs-review, ready-for-committer); platform narrows to one CI platform; limit caps the list. Cite https://commitfest.postgresql.org/patch/<cf_id>/.

Input parameters:

- `entry_status` (string): Restrict to entries in this status (e.g. needs-review, ready-for-committer). Default: any active status.
- `limit` (number): Max entries (1-500, default 100)
- `platform` (string): Restrict to entries failing on this platform keyword (e.g. linux, macos, windows, freebsd, netbsd, openbsd).

### `get_attachment` (~183 tokens)

The content of one attachment of a mail message (a patch, a log, a test case), chosen by 1-based index or a filename substring, as listed in get_message/get_thread attachments. Text attachments come back as decoded text (a .patch is ready to read or `git apply`), binary ones as base64 with encoding=base64; size is the decoded byte count. Cite the message as https://postgr.es/m/<message_id>.

Input parameters:

- `filename` (string): Substring to match against attachment filenames
- `inbox` (string): Mailing list name, e.g. pgsql-general. Defaults to pgsql-hackers when omitted. Aliases: list, inbox_name.
- `index` (number): 1-based attachment index (from get_message attachment metadata)
- `message_id` (string, required): Message-ID (with or without angle brackets)

### `get_callers` (~134 tokens)

Reverse call graph: the symbols that call or reference qualified_name (from search_symbols), each with its file and lines, like `git grep -w <name>` restricted to call sites. HEAD only; macro expansion and function pointers are not followed. limit caps the list. Read a caller with git_show_file at its line.

Input parameters:

- `limit` (number): Maximum results (1-100, default 50)
- `qualified_name` (string, required): Fully qualified symbol name
- `repository` (string): Repository name, e.g. pgbouncer. Defaults to postgres when omitted. Aliases: repo, repo_name.

### `get_contributor` (~135 tokens)

A contributor's profile by email (exact) or name (substring): canonical name, page hash, totals and per-release counts by role (authored, reviewed, committed, reported, ...), the 5 latest credited commits with sha, and buildfarm animals they own. People who hid their page are excluded from name lookups. For what they did recently use whats_new. Cite https://pg.ddx.io/community/<hash>.

Input parameters:

- `email` (string): Exact :person/canonical-email match
- `name` (string): :person/canonical-name substring (case-insensitive). Used only when email is not provided.

### `get_doc_page` (~103 tokens)

One PostgreSQL documentation page as text: path (e.g. "brin.html" or "sql-createtable.html") and optional version ("17", "current"; default the latest). Returns its url, path and stored text (body) with fetched_at. Use search_docs to find the path. Cite the returned url.

Input parameters:

- `path` (string, required): URL path of the documentation page
- `version` (string): Documentation version (default: latest available)

### `get_entry` (~98 tokens)

One commitfest entry by cf_id (the number in commitfest.postgresql.org/patch/<cf_id>/): title, status, commitfest, authors, reviewers, committer, its mail threads and branches. Read the thread with get_thread; cfbot state with find_failing_patches. Cite https://commitfest.postgresql.org/patch/<cf_id>/.

Input parameters:

- `cf_id` (number, required): commitfest.postgresql.org entry id

### `get_message` (~222 tokens)

One mail message by Message-ID: headers, in_reply_to/references, decoded body text (by default the author's own words: quotes, signature and inline patch removed; full_text=true for everything, quotes included), attachments [{index, filename, content_type, size}] and has_patch. Never raw MIME: get_attachment returns an attachment's content, get_raw_message the RFC 822 source. Finds the list itself unless inbox is given. Cite as https://postgr.es/m/<message_id>.

Input parameters:

- `body_only` (boolean): Older spelling of full_text=true: the whole decoded text (default false).
- `full_text` (boolean): Return the whole decoded text, quoted replies and signature included (default false: the author's own words only).
- `inbox` (string): Mailing list name, e.g. pgsql-general. Defaults to pgsql-hackers when omitted. Aliases: list, inbox_name.
- `include_body` (boolean): Include message body (default true)
- `message_id` (string, required): Message-ID (with or without angle brackets)

### `get_thread` (~323 tokens)

A mail thread, found by any Message-ID in it, oldest first. Each message: from, to, cc, date, message_id, decoded body text (by default the author's own words: quoted replies, signature and inline patch removed; full_text=true for the whole decoded text; never raw MIME), attachments [{index, filename, content_type, size}] and has_patch. Paged: 20 messages per call (limit max 500; offset/next_offset/has_more); bodies capped at max_body_chars (default 4000; 0 = no cap; body_truncated marks a cut). include_bodies=false gives a cheap outline. Finds the list itself unless inbox is given. Cite as https://postgr.es/m/<message_id>.

Input parameters:

- `full_text` (boolean): Whole decoded text of each message, quotes included (default false: own words only).
- `inbox` (string): Mailing list name, e.g. pgsql-general. Defaults to pgsql-hackers when omitted. Aliases: list, inbox_name.
- `include_bodies` (boolean): Include message bodies (default true). false returns an outline: headers only.
- `limit` (number): Messages per page (1-500, default 20).
- `max_body_chars` (number): Cap each body at this many characters (default 4000; 0 = no cap).
- `message_id` (string, required): Message-ID of any message in the thread
- `offset` (number): Skip this many messages; use next_offset from the previous page.

### `get_wiki_page` (~80 tokens)

One wiki.postgresql.org page by title: its latest revision content with author and date. Titles are as on the wiki ("Don't Do This", "Todo"). Use search_wiki to find a title, get_wiki_history for older revisions. Cite https://wiki.postgresql.org/wiki/<title>.

Input parameters:

- `title` (string, required): Wiki page title

### `git_log` (~285 tokens)

Commit history like `git log`, newest first, filtered by path (commits touching a file), author (substring), since/until (YYYY-MM-DD or RFC 3339) and branch (any branch, tag or major version such as 17 or 9.6; default the indexed default branch). Each commit: sha (40-hex, cite it), commit_id, message (subject for a branch log), author and dates. limit default 50, max 200; offset pages; has_more. repository defaults to postgres.

Input parameters:

- `author` (string): Filter by author name or email (substring match)
- `branch` (string): Restrict the log to commits reachable from this branch (e.g. cf/5556). Default: the repository's default branch.
- `limit` (number): Maximum commits to return (1-200, default 50)
- `offset` (number): Offset for pagination (default 0)
- `path` (string): Filter commits to those touching this file path
- `repository` (string): Repository name, e.g. pgbouncer. Defaults to postgres when omitted. Aliases: repo, repo_name.
- `since` (string): Only commits after this date (YYYY-MM-DD or RFC3339)
- `until` (string): Only commits before this date (YYYY-MM-DD or RFC3339)

### `git_search` (~222 tokens)

Search postgres commit messages (or file paths with search_in=paths) by keywords, newest first (all words must appear, case-insensitive; like `git log -i --grep` with every word). Each commit: sha (40-hex, cite it), commit_id, message, author, date, people credited, and mail_refs: up to 5 messages that refer to it {inbox, message_id, subject, url, how_it_refers: discussion-trailer, committers-announcement or mentions-hash}. repository defaults to postgres; limit default 50, max 200. Details with commit_history.

Input parameters:

- `limit` (number): Maximum results to return (1-200, default 50)
- `query` (string, required): Search query string
- `repository` (string): Repository name, e.g. pgbouncer. Defaults to postgres when omitted. Aliases: repo, repo_name.
- `search_in` (string): What to search: "messages" (commit messages, default) or "paths" (file paths)

### `hybrid_search` (~284 tokens)

Mailing-list search, keyword (BM25 over the authors' own words) and semantic (embeddings) blended per thread, over every list unless inbox names one or more. Returns up to limit threads (default 20, max 50), best first, each its best message: inbox, message_id, subject, from, date, thread_id, fused_score (0-1), similarity and a short excerpt of the author's own words. Like a ranked `git log --grep` for mail. Cite as https://postgr.es/m/<message_id>; read with get_thread or get_message (inbox from the result).

Input parameters:

- `inbox` (string): Lists to search: omit (or "all") for every list, the default; one name such as pgsql-bugs; or several, comma-separated ("pgsql-hackers,pgsql-bugs"). Names from list_inboxes. Aliases: list, inbox_name.
- `keyword_weight` (number): Weight for keyword results in the fusion, 0.0-1.0 (default 0.65). Raise it for exact identifiers (function names, GUCs, error text); lower it to lean on semantic similarity.
- `limit` (number): Maximum results to return (1-50, default 20)
- `query` (string, required): Search query (supports both keywords and natural language)

### `list_sources` (~158 tokens)

Every source behind this server with its kind (inbox, repository, docs, wiki, discord), item count, last update and freshness (fresh/stale): the 56 mailing lists, the git repositories, the documentation releases, the wiki and Discord. Use it to learn the inbox and repository names other tools accept and how current each is. kind filters; min_items drops small sources.

Input parameters:

- `kind` (string): Filter to a single kind (inbox|repository|docs|wiki|discord)
- `limit` (number): Forward-compat: max rows (default unlimited)
- `min_items` (number): Only return sources with at least this many items
- `offset` (number): Forward-compat: skip the first N rows

### `search_commitfest` (~226 tokens)

Full-text search of commitfest entries (title first, then author/reviewer/committer names), in the current commitfest by default, or commitfests="all" (back to 2014) or a comma list of numbers. Websearch syntax: "phrase", -term, or. Each result: cf_id, title, status, commitfest(s), people, last update, the commitfest.postgresql.org patch url and the pg.ddx.io thread link. limit default 20, max 50. Details with get_entry; cite the patch url.

Input parameters:

- `commitfests` (string): "current" (default: the commitfest commitfest.postgresql.org/current/ shows), "all", or a comma list of commitfest numbers, e.g. "62,61" or "current,55"; see /api/v2/commitfest/commitfests
- `limit` (integer): Max results (default 20, max 50)
- `query` (string, required): Search query, e.g. NUMA or "Tom Lane"

### `search` (~263 tokens)

Keyword search of mailing-list messages (BM25 over subject, sender and the authors' own words), every list unless inbox narrows it. Prefixes: s:subject f:from t:to d:2024-01..2025-01 b:body. Returns messages best first (limit default 50, max 200; offset pages; collapse_threads=true keeps one per thread), each with inbox, message_id, subject, from, date, thread_id and an excerpt, plus total and has_more. Use it for exact words, names and identifiers; hybrid_search for questions. Cite as https://postgr.es/m/<message_id>.

Input parameters:

- `collapse_threads` (boolean): Deduplicate results to one per thread, keeping the highest-scoring hit (default false)
- `inbox` (string): Lists to search: omit (or "all") for every list, the default; one name such as pgsql-bugs; or several, comma-separated ("pgsql-hackers,pgsql-bugs"). Names from list_inboxes. Aliases: list, inbox_name.
- `limit` (number): Maximum results to return (1-200, default 50)
- `offset` (number): Offset for pagination
- `query` (string, required): Search query string

### `search_docs` (~200 tokens)

Full-text search of the PostgreSQL documentation (websearch syntax: "phrase", -term, or), title matches first. Searches the current release unless version (one, e.g. "17") or versions (several) is given; a page matching in several selected versions is one result listing each version and its URL. Returns title, path, url, the versions it matched and a snippet. Read a page with get_doc_page; cite its postgresql.org/docs url.

Input parameters:

- `limit` (number): Maximum results to return (1-50, default 20)
- `query` (string, required): Search query string
- `version` (string): One documentation version (e.g. "18", "17"; "current" = the current release). Ignored when versions is given.
- `versions` (array): Documentation versions to search, e.g. ["18", "17"] (default: the current release; see list_doc_versions)

### `search_symbols` (~229 tokens)

Find C functions, macros, types and variables by name in the indexed source (HEAD): BM25 on names, then fuzzy and body matches. Filter by kind (function, struct, macro, ...), language, repository. Each result: qualified_name, name, kind, language, file, start_line and end_line. Like `ctags`/`git grep -w` for definitions. Follow with get_callers, or git_show_file with start_line/end_line to read the body.

Input parameters:

- `kind` (string): Filter by symbol kind: function, class, method, struct, enum, interface, trait, etc.
- `language` (string): Filter by language: go, python, c, cpp, rust, java, typescript, javascript, etc.
- `limit` (number): Maximum results (1-200, default 50)
- `query` (string, required): Search query (symbol name, partial name, or keyword)
- `repository` (string): Repository name, e.g. pgbouncer. Defaults to postgres when omitted. Aliases: repo, repo_name.

### `search_wiki` (~191 tokens)

Full-text search (websearch syntax) of wiki.postgresql.org. By default searches each page's current text; history=true searches every revision and returns, per page, the latest matching revision (rev_id, revision_url, author, comment, timestamp, is_current) and how many matched; since/until bound the revision dates. Read a page with get_wiki_page; cite its wiki url.

Input parameters:

- `history` (boolean): Search every revision, not only the current text (default false)
- `limit` (number): Maximum results to return (1-50, default 20)
- `query` (string, required): Search query string
- `since` (string): With history: only revisions at or after this date (YYYY-MM-DD or ISO-8601)
- `until` (string): With history: only revisions before the end of this date (YYYY-MM-DD or ISO-8601)

### `cross_reference` (~314 tokens)

What relates to one stored item. ref: a Message-ID or message URL, a commit SHA, repo@sha or commit URL, a docs URL or page.html, a wiki URL or title, a commitfest URL, or CVE-YYYY-NNNNN. Returns the related mail threads, commits, docs pages, wiki pages, commitfest entries and people, each listing why, strongest first: explicit links (Discussion: trailers, messages citing the SHA, commitfest thread links, pages linking the URL, commits to a docs page's SGML), shared identifiers (BUG #N, CVE ids, code identifiers), then similar topic by embeddings. kinds narrows; limit_per_kind (default 5, max 25); version picks the docs release. ~3 s budget: late kinds go in timed_out with partial=true. Cite each item's url. Same as GET /api/v2/related.

Input parameters:

- `kinds` (array): Result kinds to return (default all)
- `limit_per_kind` (number): Items per kind (1-25, default 5)
- `ref` (string, required): The item: e.g. 19705-548fda77321e062d@postgresql.org, https://www.postgresql.org/docs/current/brin.html, Don't Do This, b536d5a0bd7a, CVE-2024-10979
- `version` (string): Docs version for a page given without one (default current), e.g. 17

### `whats_new` (~443 tokens)

Catch up: everything a person or group did, and where they were named, in a window, newest first per kind (mail grouped by thread). who: a name (fuzzy; close matches return ambiguous [{name, hash}]), email, hash:<8 hex>, team:<slug>, org:<name>, new:<window>, a phrase ('the core team'), or a list. since/until: ISO date, today, 2w/14d/1m, pg19 (a release's development window), all; default 14 days. during: a calendar event uid/url. topic narrows by hybrid search. Kinds: waiting (one person, no topic: what is on their radar, each with group, priority 0-100 and why), mail_from, mail_to_cc, mail_mention, git_author, git_credit, git_mention, wiki_edit, wiki_mention, commitfest, calendar, roster. Names, page hashes and urls only, never addresses. ~4 s budget; late kinds go in timed_out with partial=true. Cite each item's url. Same as GET /api/v2/activity.

Input parameters:

- `during` (string): A calendar event uid or url, or pg19: sets since and until to its dates
- `kinds` (array): Result kinds to return (default all)
- `limit` (number): Items per kind (1-100, default 20)
- `since` (string): Window start: 2026-10-01, 2w, 14d, 1m, today, pg19, all (default 14 days ago)
- `topic` (string): Optional topic that narrows every kind, e.g. NUMA, benchmarking
- `until` (string): Window end: now (default), today 08:00, 2026-10-21T08:00, pg19
- `who` (required): Who: 'Tom Lane', 'team:core-team', 'org:Amazon/AWS', 'new:30d', a 8-hex person hash, an email; a string or a list of strings

### `code_search` (~516 tokens)

Search the postgres source at any commit, branch, tag or date (or across history), from the code-history index: every line of every commit, byte-exact. A literal by default; match=regex for a PostgreSQL regex (^ and $ are per line; \b \< \> as in grep), match=identifier for any of snake_case/CamelCase/UPPER spellings. File sets agree with `git grep -l` at the same commit. Results name the resolved sha and ref. complete=false means the totals are lower bounds (cut says why). With history (all | <rev> | A..B) it returns every matching line any commit in the range had, with the commit that added it and the one that removed it.

Input parameters:

- `budget_ms` (number): Time budget (100-15000, default 10000)
- `case_sensitive` (boolean): Default true
- `context` (number): Lines of context around each match (0-20, default 0)
- `exclude` (array): Path globs to drop
- `history` (string): Search history instead of one commit: all, a rev (everything up to it on its line), or A..B (B's line after A, as git log A..B)
- `include` (array): Path globs to keep, e.g. *.c, src/include/**/*.h
- `limit` (number): Results per page (1-2000, default 100)
- `match` (string): How to read query (default literal)
- `offset` (number): Results to skip (paging)
- `output` (string): lines (default): each matching line with its number; files: paths with match counts; counts: [path, count] pairs
- `paths` (array): Directories or globs to search under, e.g. src/backend/storage
- `query` (string, required): Text, regex or identifier to find
- `repository` (string): Repository; only postgres is in the code-history index for now (the default).
- `version` (string): Commit to read: a SHA or 7+ hex prefix, a branch (master, REL_17_STABLE), a tag (REL_17_0), a major (17, 9.6), a release (17.2), <ref>@YYYY-MM-DD (the branch's last commit by that day), or <ref>~N. D…

### `read_file` (~233 tokens)

Read a file of the postgres source at any commit, branch, tag or date from the code-history index: numbered lines, at most 2000 per call (page with start_line/end_line), byte-exact to git. Says total_lines and complete; binary files return their metadata only.

Input parameters:

- `end_line` (number): Last line (default start_line + 1999)
- `path` (string, required): Path from the repository root, e.g. src/backend/access/transam/xlog.c
- `repository` (string): Repository; only postgres is in the code-history index for now (the default).
- `start_line` (number): First line (default 1)
- `version` (string): Commit to read: a SHA or 7+ hex prefix, a branch (master, REL_17_STABLE), a tag (REL_17_0), a major (17, 9.6), a release (17.2), <ref>@YYYY-MM-DD (the branch's last commit by that day), or <ref>~N. D…

### `list_files` (~225 tokens)

List the files of the postgres source under a directory at any commit, branch, tag or date (code-history index), with line counts; depth folds deeper paths into directories with file counts; glob filters paths.

Input parameters:

- `depth` (number): Levels below path to list (default all)
- `glob` (string): Path glob, e.g. *.dat or src/include/catalog/*.h
- `limit` (number): Entries (1-5000, default 500)
- `path` (string): Directory (default the root)
- `repository` (string): Repository; only postgres is in the code-history index for now (the default).
- `version` (string): Commit to read: a SHA or 7+ hex prefix, a branch (master, REL_17_STABLE), a tag (REL_17_0), a major (17, 9.6), a release (17.2), <ref>@YYYY-MM-DD (the branch's last commit by that day), or <ref>~N. D…

### `blame` (~317 tokens)

Per-line blame of a postgres file at any commit (code-history index): the commit that wrote each line, as `git blame` with git's default diff. ignore_revs (default true) passes through the pgindent/reformat commits of .git-blame-ignore-revs when they replaced lines one-for-one; reverse adds the commit that removed each line later on the version's branch. At most 2000 lines per call; commits lists author, date and subject per short sha. Not the same as git_blame, which returns a file's commit history.

Input parameters:

- `end_line` (number): Last line (default start_line + 1999)
- `ignore_revs` (boolean): Pass through .git-blame-ignore-revs commits (default true)
- `path` (string, required): Path from the repository root
- `repository` (string): Repository; only postgres is in the code-history index for now (the default).
- `reverse` (boolean): Also name the commit that later removed each line (default false)
- `start_line` (number): First line (default 1)
- `version` (string): Commit to read: a SHA or 7+ hex prefix, a branch (master, REL_17_STABLE), a tag (REL_17_0), a major (17, 9.6), a release (17.2), <ref>@YYYY-MM-DD (the branch's last commit by that day), or <ref>~N. D…

### `get_commit` (~199 tokens)

One postgres commit from the code-history index: message, parsed trailers (names only), Discussion threads (links to the archived thread on /m/), back-patch siblings on other branches, the first release tag on each branch that contains it, files changed with line counts, and its diff rendered from the index (at most 3000 lines; diff_complete says whether it was cut). Accepts a SHA (7+ hex), branch, tag or any version form.

Input parameters:

- `commit` (string, required): SHA or prefix (7+ hex), or any version form (REL_17_0, master~2)
- `context` (number): Diff context lines (0-1000, default 3)
- `diff` (boolean): Include the diff (default true)
- `paths` (array): Only diff files under these paths
- `repository` (string): Repository; only postgres is in the code-history index for now (the default).

### `search_commits` (~319 tokens)

Search postgres commits (code-history index), newest first, like git log: range (a rev, or A..B; default the newest release branch), message (regex, case-insensitive), author (regex over author, committer and Author/Co-authored-by/Reviewed-by trailers), paths, since/until (YYYY-MM-DD), changed_lines (a regex an added or removed line matches, as git log -G) and occurrences (a string whose count in a file changed, as git log -S).

Input parameters:

- `author` (string): Regex over author, committer and credit trailers
- `budget_ms` (number): Time budget (100-15000, default 10000)
- `changed_lines` (string): Regex an added or removed line matches (git log -G)
- `limit` (number): Commits (1-200, default 30)
- `message` (string): Regex over the commit message
- `occurrences` (string): String whose number of occurrences changed (git log -S)
- `offset` (number): Commits to skip
- `paths` (array): Path prefixes the commit touched
- `range` (string): A rev (its line back to the root) or A..B; default the newest release branch
- `repository` (string): Repository; only postgres is in the code-history index for now (the default).
- `since` (string): YYYY-MM-DD (committed on or after)
- `until` (string): YYYY-MM-DD (committed before)

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/io-ddx-pg-archive/pg-2#diagnostics

## Score history

- 2026-10-24: 68
- 2026-10-23: 67

## Common questions

### What is the pg.ddx.io: the PostgreSQL community's memory MCP server?

pg.ddx.io: the PostgreSQL community's memory is an MCP server listed in the public MCP registry as io.ddx.pg/archive. Every PostgreSQL mailing list since 1991, linked to the commits, code, commitfest and docs. This page covers its hosted endpoint (https://pg.ddx.io/mcp/sse).

### Is the pg.ddx.io: the PostgreSQL community's memory MCP server safe to use?

pg.ddx.io: the PostgreSQL community's memory scores 68 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the pg.ddx.io: the PostgreSQL community's memory MCP server expose?

pg.ddx.io: the PostgreSQL community's memory exposes 29 tools: retrieve_context, commit_buildfarm_status, commit_history, find_failing_patches, get_attachment, and 24 more. Their descriptions and schemas cost roughly 6,632 tokens of context every time the server is loaded.

### Does the pg.ddx.io: the PostgreSQL community's memory MCP server require authentication?

No. We connected to pg.ddx.io: the PostgreSQL community's memory without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

### Is the pg.ddx.io: the PostgreSQL community's memory MCP server still maintained?

pg.ddx.io: the PostgreSQL community's memory is still listed as active in the MCP registry. We last reached this channel on 24 October 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

## Links

- Remote endpoint: https://pg.ddx.io/mcp/sse
- Website: https://pg.ddx.io/
- Changelog RSS feed: https://verifymcp.io/servers/io-ddx-pg-archive/pg-2.xml
- Changelog JSON feed: https://verifymcp.io/servers/io-ddx-pg-archive/pg-2.json
- HTML version of this page: https://verifymcp.io/servers/io-ddx-pg-archive/pg-2