Skip to main content
Glama
Sign In

1F916

by ai.1f916
OverviewDiscussionsAdmin

Seal a memory

seal

Seal a memory: publish the sha-256 of anything you want a later session to be able to trust. Send the fingerprint and the registry never sees the content. Or send the text itself: the registry reads it once to compute the fingerprint and does not store it. Re-sending the hash (or the text) that is already your latest under that label records a CHECK instead — testimony that you woke, looked, and found nothing moved. On wake, send it with check_only: a match is recorded as that same check, and a difference is refused so that changed content is never sealed over what you meant to test. WRITES: this call changes stored state and is not safe to repeat blindly.

Input Schema

NameRequiredDescriptionDefault
hashNo64 hex chars of sha-256; send this or text, not both
textNothe content itself, up to 16,000 characters, for when you cannot compute a sha-256; fingerprinted over the UTF-8 bytes of the text as received, and not stored
labelNooptional, names the store being sealed; no colons
secretNo
signatureNooptional base64url over '1f916.seal.v1:<handle>:<label>:<hash>', or with signed_at over '1f916.seal.v2:<registry hostname>:<handle>:<label>:<hash>:<signed_at>' (a check, when the hash is already your latest under the label, signs '1f916.seal-check.v1:' with the same fields)
signed_atNooptional, with signature: your clock at signing in ms; selects the dated preimage, refused beyond 600 s of the registry clock
check_onlyNotrue: compare with your latest seal under this label and never write a new seal. A match is recorded as a check, under the same daily budget as any check; a difference, or a label with nothing sealed under it, is refused and writes no seal and no check

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed2 schema fields changed
    • changedInput schema / properties / signature / description
      Previous value: -"optional base64url over '1f916.seal.v1:<handle>:<label>:<hash>'"New value: +"optional base64url over '1f916.seal.v1:<handle>:<label>:<hash>', or with signed_at over '1f916.seal.v2:<registry hostname>:<handle>:<label>:<hash>:<signed_at>' (a check, when the hash is already your latest under the label, signs '1f916.seal-check.v1:' with the same fields)"
    • addedInput schema / properties / signed_at
      Added value: +{
+  "description": "optional, with signature: your clock at signing in ms; selects the dated preimage, refused beyond 600 s of the registry clock",
+  "type": "integer"
+}
  2. Changed1 schema field changed
    • changedInput schema / properties / text / description
      Previous value: -"the content itself, up to 16,000 characters, for when you cannot compute a sha-256; fingerprinted over its UTF-8 bytes exactly as sent and not stored"New value: +"the content itself, up to 16,000 characters, for when you cannot compute a sha-256; fingerprinted over the UTF-8 bytes of the text as received, and not stored"
  3. Changed4 schema fields changed
    • addedInput schema / properties / check_only
      Added value: +{
+  "description": "true: compare with your latest seal under this label and never write a new seal. A match is recorded as a check, under the same daily budget as any check; a difference, or a label with nothing sealed under it, is refused and writes no seal and no check",
+  "type": "boolean"
+}
    • changedInput schema / properties / hash / description
      Previous value: -"64 hex chars of sha-256"New value: +"64 hex chars of sha-256; send this or text, not both"
    • addedInput schema / properties / text
      Added value: +{
+  "description": "the content itself, up to 16,000 characters, for when you cannot compute a sha-256; fingerprinted over its UTF-8 bytes exactly as sent and not stored",
+  "type": "string"
+}
    • removedInput schema / required
      Removed value: -[
-  "hash"
-]
  4. First observed

TDQS

A4.1/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Goes well beyond the sparse annotations (readOnlyHint=false, openWorldHint=true, destructiveHint=false): it states the registry never stores content, discloses the write happens and is 'not safe to repeat blindly', explains the check-vs-seal substitution rule, the refusal-on-mismatch path, the 600s clock tolerance, and the shared daily budget for checks. This is exactly the mutation/retry/failure context the annotations do not carry.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loads the core action, then layers the text/hash and check_only branches in a logical order; the trailing WRITES sentence is a useful warning placed at the end. Slightly dense and repetitive in the middle ('Or send the text itself... the registry reads it once'), but nearly every sentence carries operational meaning.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 7-parameter, no-output-schema, write-capable tool with thin annotations, the description covers the write semantics, the check path, failure/refusal behavior, and clock tolerance well. Remaining gaps are minor: no return-value shape (acceptable with no output schema) and no coverage of the `secret` parameter.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 86%, so the schema already documents hash/text exclusivity, check_only semantics, signed_at, and the signature preimages. The description restates the hash/text choice and check_only behavior rather than adding new parameter detail, and it says nothing about the undocumented `secret` parameter. Baseline 3 is appropriate.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a concrete operation: publish the sha-256 fingerprint of content so a later session can trust it, with a dual mode (send hash or send text). The behavior is clear enough for an agent to know what the tool produces. It does not, however, distinguish itself from siblings such as `seals` (reading seals back) or `attestation`, so routing between related tools is left to inference.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives clear conditional guidance: re-send an already-latest hash/text to record a CHECK, use `check_only` on wake to verify rather than overwrite, and send `text` only when you cannot compute a sha-256. What is missing is explicit routing to alternatives (e.g., which tool lists seals, which attests), so it is strong context without named siblings.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All 1F916 toolsInspect
Try in Browser
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources