Engram
OCI · GHCR.IO/RWNALDS/ENGRAM:0.1.2 · SCANNED SEP 20
Git-backed markdown memory for AI agents; search hides superseded and archived notes.
Available components
How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. How we score → Why this is hard to score →
Supply Chain Security0
- Malware scan not yet available for this package.Unverified
- Known CVEs could not be checked: this artifact ships no SBOM, so there is no dependency list to read. Publishing one would let us assess it.Unverified
- Install-script risk not yet assessed.Unverified
- Dependency health could not be checked: this artifact ships no SBOM, so there is no dependency list to read. Publishing one would let us assess it.Unverified
Provenance & Transparency35
- Source repository is publicly reachable at the declared URL. View diagnostics → Pass
- Provenance check failed: no build-provenance attestation is published. See how to fix → View diagnostics → Fail
- License check failed: the license (NOASSERTION) isn't a recognized OSI-approved license. See how to fix → Fail
- Actively maintained (last published 130 days ago).Pass
- Publishes a security disclosure policy (SECURITY.md).Pass
Schema Quality & AI Usability73
- AI-judged instruction clarity (good).Pass
- Context-footprint check failed: tool/resource definitions use about 2026 tokens (~126/item across 16 items; 16 tools + 0 resources), over budget; trim descriptions and params. See how to fix → Fail
- Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management80
- Stability observed for 24 of 30 days with no destabilising changes; credit accrues until the full window elapses.Partial
Tool Coverage100
- 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
- 100% of tool parameters carry a description.Pass
Tool Safety75
- No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
- 0 of 2 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation; "brain_edit" implies "overwrite" and declares no destructiveHint at all, which the MCP spec reads as destructive by default. See how to fix → Fail
- An AI judge read all 16 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
- Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.Pass
Unverified: 1 category
A category scored 0 because we could not verify it: a data source with nothing on this package, evidence we could not reach, or a check we could not run. We only credit what we can confirm.
How do I install the Engram MCP server?
Engram runs locally as a container image, launched with docker run --rm -i ghcr.io/rwnalds/engram:0.1.2. Ready-made configuration for Claude, Cursor, VS Code, Codex and 3 more is on this page, copied from each client's own documentation.
oci · ghcr.io/rwnalds/engram:0.1.2
claude mcp add rwnalds-engram -- docker run --rm -i ghcr.io/rwnalds/engram:0.1.2
{
"mcpServers": {
"rwnalds-engram": {
"command": "docker",
"args": [
"run",
"--rm",
"-i",
"ghcr.io/rwnalds/engram:0.1.2"
]
}
}
} {
"servers": {
"rwnalds-engram": {
"command": "docker",
"args": [
"run",
"--rm",
"-i",
"ghcr.io/rwnalds/engram:0.1.2"
]
}
}
} codex mcp add rwnalds-engram -- docker run --rm -i ghcr.io/rwnalds/engram:0.1.2
{
"$schema": "https://opencode.ai/config.json",
"mcp": {
"rwnalds-engram": {
"type": "local",
"command": [
"docker",
"run",
"--rm",
"-i",
"ghcr.io/rwnalds/engram:0.1.2"
],
"enabled": true
}
}
} mcp_servers:
rwnalds-engram:
command: "docker"
args: ["run", "--rm", "-i", "ghcr.io/rwnalds/engram:0.1.2"] {
"McpServers": {
"rwnalds-engram": {
"Transport": "stdio",
"Command": "docker",
"Arguments": [
"run",
"--rm",
"-i",
"ghcr.io/rwnalds/engram:0.1.2"
]
}
}
} {
"mcpServers": {
"rwnalds-engram": {
"command": "docker",
"args": [
"run",
"--rm",
"-i",
"ghcr.io/rwnalds/engram:0.1.2"
]
}
}
} Every change we have recorded for this component, newest first. Security-relevant changes are always shown. ▲ marks a change for the better, ▼ a change for the worse; unmarked changes are neutral.
- 20 Sept 26 −3
- Stability: pass → 0.80 functional
- 19 Sept 26 +1
- Stability: 0.97 → pass security
- 17 Sept 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 90 to 93. That category is still filling its 30-day observation window: 27 days of observed history at the previous scan, 28 at this one. The score rises as the window fills, whether or not the server changes.
- 15 Sept 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 83 to 87. That category is still filling its 30-day observation window: 25 days of observed history at the previous scan, 26 at this one. The score rises as the window fills, whether or not the server changes.
- 13 Sept 26 −3
- Stability: pass → 0.80 functional
- 12 Sept 26 +1
- Stability: 0.97 → pass security
- 10 Sept 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 90 to 93. That category is still filling its 30-day observation window: 27 days of observed history at the previous scan, 28 at this one. The score rises as the window fills, whether or not the server changes.
- 9 Sept 26 0
- Schema quality: excellent → good functional
Diagnostic detail from the automated scan of this channel: what the scanner observed at each step, so you can see exactly where a check passed or failed. It is informational only and never changes the trust score.
Captured 20 Sept 2026 · Analysed oci/ghcr.io/rwnalds/engram:0.1.2
Provenance No attestation
The registry publishes no build provenance for this version, so there is nothing to verify.
| Result | No attestation |
|---|---|
| Ecosystem | oci |
| Reason | No attestation published |
Background: How many MCP packages publish verified provenance →
The tools this component advertises to a client, with an estimated token cost for each. Expand a tool to see its parameters and schema. The per-tool counts are indicative and are not scored directly; the schema's total context footprint is one signal in Schema Quality & AI Usability. A tool's description is untrusted text the model reads on every call, which is what makes this list a security surface and not just an inventory: how tool poisoning works →
brain_append ~39
Append text to a note (creates it if missing).
| Name | Type | Req | Description |
|---|---|---|---|
| path | string | yes | vault-relative path |
| text | string | yes | text to append |
No output schema declared.
No examples provided.
brain_backlinks ~27
Notes that link to the given note.
| Name | Type | Req | Description |
|---|---|---|---|
| path | string | yes | vault-relative path |
No output schema declared.
No examples provided.
brain_capture ~121
Hand over a ROUGH note / brain-dump and let the vault file it. An agent loop reads SCHEMA.md, searches for what already exists, then deliberately creates a new note, appends to a matching one, or archives what this supersedes — and returns a manifest of every path it touched. It reads a note before overwriting it, and never deletes. Use when you have unstructured input and don't want to choose the path yourself; use brain_write when you do.
| Name | Type | Req | Description |
|---|---|---|---|
| text | string | yes | the rough note / brain dump to file |
No output schema declared.
No examples provided.
brain_create_folder ~32
Create a new folder in the vault (with a .gitkeep).
| Name | Type | Req | Description |
|---|---|---|---|
| path | string | yes | folder path |
No output schema declared.
No examples provided.
brain_delete ~68
Delete a note (recoverable via git history). Prefer brain_move into an archive folder — deleting destroys the reasoning trail, archiving only removes it from search. Delete when the note is wrong or duplicated, archive when it is merely no longer true.
| Name | Type | Req | Description |
|---|---|---|---|
| path | string | yes | vault-relative path |
No output schema declared.
No examples provided.
brain_edit ~165
Overwrite a note. Pass `content` (full raw markdown incl. frontmatter) — read first with brain_read, then write the whole file back. Also accepts `body` (+ optional `frontmatter`) like brain_write.
| Name | Type | Req | Description |
|---|---|---|---|
| allow_conflict | boolean | – | confirm a near-duplicate of a live note is deliberate — prefer brain_supersede (default false) |
| body | string | – | markdown body (alternative to content; pair with `frontmatter`) |
| content | string | – | full raw markdown incl. frontmatter |
| frontmatter | object | – | YAML frontmatter object (with `body`) |
| overwrite | boolean | – | confirm you mean to replace an existing note with much shorter content (default false) |
| path | string | yes | vault-relative path |
No output schema declared.
No examples provided.
brain_graph ~37
The knowledge graph (nodes + edges from wikilinks and related:). Optional folder filter.
| Name | Type | Req | Description |
|---|---|---|---|
| folder | string | – | optional folder filter |
No output schema declared.
No examples provided.
brain_list ~51
List every note with metadata (path, title, folder, type, tags, status, authority). Use to discover what exists. `authority` tells you which notes are source-of-truth and which are history.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
brain_move ~104
Move or rename a note. This is how you retire something: when a note stops being true, move it to an archive folder (see brain_schema → conventions.ranking.archiveFolders) rather than deleting it. Archived notes are demoted in search and excluded by default, so they stop misleading agents while the reasoning trail survives. Leave a pointer in the replacement note saying what superseded what.
| Name | Type | Req | Description |
|---|---|---|---|
| from | string | yes | current path |
| to | string | yes | new path |
No output schema declared.
No examples provided.
brain_read ~147
Read a note: full markdown (frontmatter + body), its `authority`, and its backlinks. Path is vault-relative, e.g. 'clients/mks/mks.md'. Pass `section` to read just one heading's content instead of the whole file — cheaper on long documents. If the heading isn't found you get the list of available headings back. Always check the returned `authority` before acting on the content: `archived` and `superseded` notes are history, not instructions.
| Name | Type | Req | Description |
|---|---|---|---|
| path | string | yes | vault-relative path |
| section | string | – | optional heading to extract, e.g. 'Pricing' — matches case-insensitively by prefix |
No output schema declared.
No examples provided.
brain_recent ~65
Notes changed most recently, newest first. Use to catch up on what humans or other agents have done since you last looked.
| Name | Type | Req | Description |
|---|---|---|---|
| limit | number | – | max results (default 20) |
| since | string | – | optional ISO date/time — only notes modified at or after this |
No output schema declared.
No examples provided.
brain_schema ~85
READ THIS FIRST, before searching or writing. Returns the vault's SCHEMA.md (folder taxonomy, frontmatter conventions, wikilink model, write protocol) AND a live description of this specific vault: its folders, the `status:` values actually in use, how search ranks notes by authority, and any integrity warnings. Vaults differ — never assume conventions, read them here.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
brain_search ~411
Keyword search across the vault. Returns `{ hits, excluded }`. Each hit has `path`, `title`, `folder`, `status`, `snippet`, and `authority`. IMPORTANT — ranking is by keyword relevance, NOT by truth. A superseded document repeats the query words just as often as the live one, so it can outrank it. Every hit carries an `authority`: `authoritative` (source of truth — prefer it), `current`, `provisional` (draft/proposed — never quote as settled), `superseded`, `archived`. Rank order is a suggestion; `authority` is the signal. `excluded` is the explainable-rejection list: notes that matched the query but were withheld because they are archived, superseded, or **expired** (past their `valid_until`), each with a `reason` (e.g. "superseded by price-live", "expired 2026-06-01"). **When you deliberately ignore a stale note, cite its `excluded` entry** — say what you skipped and why, instead of quoting it. If you need a withheld note, re-search with `includeInvalid: true` (superseded/expired) or `includeArchive: true`. Before quoting a price, guarantee, contract term, or any other single-valued fact, open the `authoritative` note. If two live notes disagree on such a fact, that is a defect in the vault — report it rather than averaging them.
| Name | Type | Req | Description |
|---|---|---|---|
| folder | string | – | restrict to one top-level folder, e.g. 'decisions' |
| includeArchive | boolean | – | return archived notes as hits (default false → they go to `excluded`) |
| includeInvalid | boolean | – | return superseded/expired notes as hits, still demoted (default false → `excluded`) |
| limit | number | – | max hits (default 20) |
| query | string | yes | search query — a few keywords beat a full sentence |
No output schema declared.
No examples provided.
brain_supersede ~264
Retire a fact and replace it, atomically. When a fact changes (a price, a term, a decision), DON'T just add a new note — the old value keeps matching searches and gets quoted. `brain_supersede(from, to)` marks the old note superseded in place (body preserved) and links it to the new one, in a single commit, so add-and-retire can't drift apart. After this, search withholds the old note with the reason "superseded by <to>". `from` = the note being retired; `to` = the note that replaces it. If `to` doesn't exist yet, pass `body` (its markdown) and it's created; otherwise write `to` first, then supersede. `reason` is recorded on the old note. Prefer this over brain_move for a value that changed (move is for relocation).
| Name | Type | Req | Description |
|---|---|---|---|
| body | string | – | optional — markdown for `to` if it doesn't exist yet |
| from | string | yes | the note being retired (vault-relative path) |
| reason | string | – | optional — why it was retired, e.g. 'repriced Q3' |
| to | string | yes | the note that replaces it (vault-relative path) |
No output schema declared.
No examples provided.
brain_tree ~17
Return the folder/file tree of the vault.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
brain_write ~393
Create or overwrite a note. PREFER `body` (markdown, no frontmatter) + `frontmatter` (an object): the YAML is serialised for you and always parses. Hand-writing frontmatter into `content` risks invalid YAML — a note whose frontmatter fails to parse loses its status, tags and title on every read, so a note claiming `status: locked` would rank as an ordinary one. Such a write is REJECTED. Follow SCHEMA.md: kebab-case path, dated names for daily/decisions, frontmatter with title/type/tags/status. Path vault-relative (e.g. decisions/foo-2026-07-09.md). WHEN A FACT CHANGED, DON'T CREATE A DATED SIBLING. Writing `acme-pricing-2026.md` next to a live `acme-pricing.md` leaves two live notes disagreeing about one number, and the older one usually matches the query better — this is exactly how a retired price gets quoted. Such a write is REJECTED; use brain_supersede instead, which retires the old note and adds the new one in a single commit. `allow_conflict: true` overrides it when both notes genuinely belong.
| Name | Type | Req | Description |
|---|---|---|---|
| allow_conflict | boolean | – | confirm a near-duplicate of a live note is deliberate — prefer brain_supersede (default false) |
| body | string | – | markdown body (pair with `frontmatter`) |
| content | string | – | full raw markdown incl. frontmatter — alternative to body+frontmatter |
| frontmatter | object | – | YAML frontmatter object: title, type, tags, status, related, ... |
| overwrite | boolean | – | confirm you mean to replace an existing note with much shorter content (default false) |
| path | string | yes | vault-relative path, e.g. decisions/foo-2026-07-09.md |
No output schema declared.
No examples provided.
What is the Engram MCP server?
Engram is an MCP server listed in the public MCP registry as io.github.rwnalds/engram. Git-backed markdown memory for AI agents; search hides superseded and archived notes. This page covers its container image (ghcr.io/rwnalds/engram:0.1.2).
Is the Engram MCP server safe to use?
Engram scores 48 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 Engram MCP server expose?
Engram exposes 16 tools: brain_schema, brain_search, brain_read, brain_list, brain_recent, and 11 more. Their descriptions and schemas cost roughly 2,026 tokens of context every time the server is loaded.
Is the Engram MCP server still maintained?
Engram is still listed as active in the MCP registry. We last reached this channel on 20 September 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.
What licence is the Engram MCP server under?
Engram declares the NOASSERTION licence, which is not on the OSI-approved list. Read the terms before using it at work, and note this covers the source only, not the cost of any service it calls.