io.github.Goldentrii/agent-recall
NPM · AGENT-RECALL-MCP · 2 COMPONENTS
Closes the Intelligent Distance: a correction you give one agent reaches every connected agent.
Available components
Last scanned (20 hours ago)
Scores higher than 59% of npm servers
Category Memory and Knowledge
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 Security81
- No malware found by supply-chain analysis.Pass
- CVE check failed: a known high-severity CVE affects zod 3.25.76, a direct dependency. A fixed version is available. View diagnostics → Fail
- No install/post-install scripts declared.Pass
- 65 of 204 dependencies flagged as unhealthy (2 deprecated). View diagnostics → Partial
Provenance & Transparency45
- 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
- Clear OSI-approved license (MIT).Pass
- Actively maintained (last published 5 days ago).Pass
- Disclosure check failed: no security disclosure policy was found in the source repository. See how to fix → Fail
Schema Quality & AI Usability73
- 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).Pass
- AI-judged instruction clarity (good).Pass
- Context-footprint check failed: tool/resource definitions use about 1862 tokens (~266/item across 7 items; 5 tools + 2 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 Management100
- No destabilizing schema changes in the last 30 days.Pass
Tool Coverage95
- 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
- 84% of tool parameters carry a description.Partial
Tool Safety100
- No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
- We read all 5 captured tool definition(s), and no name or description among them implies an irreversible operation.Pass
- An AI judge read all 7 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
Checks we ran (16)
- Malware scan
- Known CVEs
- Install scripts
- Dependency health
- Source repository
- Provenance
- License
- Maintenance
- Security disclosure
- Schema quality
- Stability
- Tool coverage
- Injection markers
- Destructive annotations
- Judged manipulation
- MCP protocol
Does not apply to this package (2)
- Endpoint SecurityScored on hosted endpoints only. A package has no live endpoint of its own for us to probe.
- Transport & ReachabilityScored on hosted endpoints only. A package has no live endpoint of its own for us to probe.
The score does not measure everything. What the score does not measure →
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.
- 18 Nov 26 0
- Stability: 0.97 → pass security
- 17 Nov 26 +15
- Malware scan: unverified → pass ▲ security
- Stability: pass → 0.97 functional
- 16 Nov 26 −15
- Malware scan: pass → unverified ▼ security
- Schema quality: 197 → 266 ▼ functional
- Package version: 3.5.2 → 3.5.3 functional
- 14 Nov 26 +15
- Malware scan: unverified → pass ▲ security
- 13 Nov 26 −11
- Malware scan: pass → unverified ▼ security
- Source repository: fail → pass ▲ security
- Schema quality: 130 → 197 ▼ functional
- Package version: 3.4.31 → 3.5.2 functional
- 11 Nov 26 0
- Stability: 0.97 → pass security
- 10 Nov 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 93 to 97. That category is still filling its 30-day observation window: 28 days of observed history at the previous scan, 29 at this one. The score rises as the window fills, whether or not the server changes.
- 8 Nov 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 87 to 90. That category is still filling its 30-day observation window: 26 days of observed history at the previous scan, 27 at this one. The score rises as the window fills, whether or not the server changes.
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 21 Nov 2026 · Analysed npm/agent-recall-mcp@3.5.3
Provenance No attestation
The registry publishes no build provenance for this version, so there is nothing to verify.
| Result | No attestation |
|---|---|
| Ecosystem | npm |
Background: How many MCP packages publish verified provenance →
Vulnerabilities 1 finding
| ID | CVE | Severity | Vector | Fix available |
|---|---|---|---|---|
| GHSA-mrhx-c7xc-m3w5 | CVE-2023-54404 | high | CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H | yes |
Background: What a vulnerability scan can and cannot prove →
Dependencies 204 packages
| Packages resolved | 204 |
|---|---|
| Deprecated | 2 |
| Stale | 64 |
| Tree resolution | Complete |
Background: SBOMs and build attestations, explained →
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 →
check Check Understanding ~488
[MID-SESSION — safe any time; for alignment, before risky decisions] Use when the user asks to validate understanding, verify alignment, or check if their interpretation matches the human's intent. Also call BEFORE a high-risk action — publish, deploy, delete, credential exposure, external send/message, or any other irreversible write — passing `action_description` (one sentence, what you're about to do). Returns matching corrections/rules/insights plus a `verdict`: `blocked` means an authoritative correction OVERRIDES the plan — read it before proceeding. To RECORD a durable human correction, pass `human_correction` as the STRUCTURED object {rule, why, applies_when} — a plain string is only STAGED for later review, never activated.
| Name | Type | Req | Description |
|---|---|---|---|
| action_description | string | – | What you're about to DO, one sentence — pass this before publish/deploy/delete/credential/external-send/irreversible-write actions. Returns matching corrections/rules/insights on the result's `action… |
| assumptions | array | – | Key assumptions you're making. |
| confidence | string | – | How confident you are. Defaults to medium. |
| decision_id | string | – | Link multiple check calls to the same decision. Auto-generated if not provided. |
| delta | string | – | The gap between your understanding and reality (or 'none'). |
| evidence | array | – | Evidence collected since prior. Each entry shifts probability. |
| goal | string | – | The goal or decision question you're checking alignment on. Required for alignment checks; optional when recording a pure decision trail (prior/posterior/evidence). |
| human_correction | – | – | After human responds: what they actually wanted. Pass the OBJECT form {rule, why, applies_when} to activate a durable correction; a plain string is only staged for review. |
| outcome | string | – | Final decision result: 'confirmed', 'rejected', 'partial', or free text. Triggers decision trail persistence. |
| posterior | number | – | Updated probability after considering evidence (0-1). |
| prior | number | – | Initial probability estimate (0-1). Start of Bayesian decision trail. |
| project | string | – | – |
| understanding | string | – | Alias for goal — use when saying 'check my understanding: X'. Provide either goal or understanding. |
No output schema declared.
No examples provided.
recall Recall ~294
[RETRIEVE — use freely, any time] Use when the user asks to recall, search, find, or look up previous memory, context, or decisions. Read-only: the call leaves the memory store byte-identical (disclosed exceptions: the hosted server's per-credential activity record). Usage counters and any feedback ratings are queued and recorded after the call: before the next write tool (waits at most ~1.5 s), by a 30 s background flush, or when the stdio server's stdin closes or it receives SIGTERM/SIGINT. A crash (kill -9, power loss) before then loses them.
| Name | Type | Req | Description |
|---|---|---|---|
| feedback | array | – | Rate previous recall results to improve future ranking. Applied AFTER this call (queued; flushed before the next write tool, by the 30 s background flush, or on stdin close or SIGTERM/SIGINT): it nev… |
| limit | integer | – | Max results after RRF merge. |
| project | string | – | – |
| query | string | yes | What to search for. |
| since | string | – | ISO date ("2026-05-01") or relative duration ("7d"). Filters journal results. |
No output schema declared.
No examples provided.
remember Remember ~187
[MID-SESSION WRITE — single fact/decision; saying it is not saving it] Use when the user asks to remember, store, note, or save a specific decision, fact, or insight.
| Name | Type | Req | Description |
|---|---|---|---|
| content | string | yes | What to remember. |
| context | string | – | Routing hint. Values: 'architecture' or 'decision' → palace/architecture room. 'blocker' or 'blocked' → palace/blockers room. 'goal' → palace/goals room. 'lesson' or 'insight' → awareness. 'qa' or 'c… |
| project | string | – | – |
No output schema declared.
No examples provided.
session_end End Session ~368
[ON SAVE/EXIT — YOU must call this; nothing auto-saves] Use when the user asks to save, checkpoint, summarize, end, retain, or persist the current session. Optionally pass close_phase / open_phase to update the project pipeline narrative spine in the same call.
| Name | Type | Req | Description |
|---|---|---|---|
| close_phase | object | – | Close the currently active pipeline phase as part of this save. Provide all three reflection fields explicitly — never auto-generated. |
| insights | array | – | Insights learned this session. |
| open_phase | object | – | Open a new pipeline phase as part of this save (e.g. when a watershed session pivots into the next strategic direction). |
| project | string | – | – |
| state | object | – | Project state card, REWRITTEN every save and loaded by the next session_start. Omitted fields carry forward; [] clears a list. Soft caps (overflow truncated with … and listed in state_card.truncated)… |
| summary | string | yes | What happened this session. Simple session: 2-3 sentences. Multi-phase session: one paragraph per completed phase (e.g. 'Phase 1 — Name: what happened. Phase 2 — Name: what happened. Decisions: X. Bl… |
| trajectory | string | – | Where is the work heading next. |
No output schema declared.
No examples provided.
session_start Start Session ~328
[ENTRY — call FIRST, before acting] Use when the user asks to start, load, continue, resume, or open memory for a project. Default mode='cards' loads the owner card, the HARD RULES and the project state card and nothing else; mode='full' adds continuity, insights, rooms, captures and the rest (and is the default when verbose=true, which promises the full JSON context); mode='lite' is a ≤500-token sketch (pull memory on demand via recall()). Read-only: the call leaves the memory store byte-identical (disclosed exceptions: AR_AB_ENABLED=1 writes an A/B arm row, and the hosted server's per-credential activity record). Usage counters and telemetry are queued and recorded before the next write tool (waits at most ~1.5 s), by a 30 s background flush, or when the stdio server's stdin closes or it receives SIGTERM/SIGINT; a crash (kill -9, power loss) before then loses them.
| Name | Type | Req | Description |
|---|---|---|---|
| context | string | – | Optional context for matching cross-project insights |
| mode | string | – | Default 'cards' = owner card + HARD RULES (P0s and behaviour policies) + state card + pending-review count, nothing else; 'full' = every section (continuity, insights, rooms, captures, …) and the def… |
| project | string | – | – |
| verbose | boolean | – | Set true to get full JSON context instead of terse summary |
No output schema declared.
No examples provided.
How do I install the io.github.Goldentrii/agent-recall MCP server?
io.github.Goldentrii/agent-recall runs locally as an npm package, launched with npx -y agent-recall-mcp. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.
npm · agent-recall-mcp
claude mcp add goldentrii-agent-recall -- npx -y agent-recall-mcp
What is the io.github.Goldentrii/agent-recall MCP server?
io.github.Goldentrii/agent-recall is an MCP server listed in the public MCP registry as io.github.Goldentrii/agent-recall. Closes the Intelligent Distance: a correction you give one agent reaches every connected agent. This page covers its npm package (agent-recall-mcp).
Is the io.github.Goldentrii/agent-recall MCP server safe to use?
io.github.Goldentrii/agent-recall scores 78 out of 100 on VerifyMCP. We recorded 1 known advisory against it as of 21 November 2026. It declares no install or post-install scripts. 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 io.github.Goldentrii/agent-recall MCP server expose?
io.github.Goldentrii/agent-recall exposes 5 tools: session_start, session_end, remember, recall, check. Their descriptions and schemas cost roughly 1,665 tokens of context every time the server is loaded.
Is the io.github.Goldentrii/agent-recall MCP server still maintained?
io.github.Goldentrii/agent-recall is still listed as active in the MCP registry. We last reached this channel on 21 November 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 io.github.Goldentrii/agent-recall MCP server under?
io.github.Goldentrii/agent-recall declares the MIT licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.
What did VerifyMCP test on the io.github.Goldentrii/agent-recall MCP server?
VerifyMCP ran 16 checks on io.github.Goldentrii/agent-recall, covering Malware scan, Known CVEs, Install scripts, Dependency health and 12 more. Endpoint Security and Transport & Reachability do not apply to an npm package, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.