aziel-runtime

Observe a memory

memory_observe

Append a new fact to the ChainLock learn chain before resolving or calibrating outcomes. Records observations additively; requires confirmation or a dry-run preview.

Instructions

Append the first memory_observation to the ChainLock learn chain (AKM-TRIAD-1.0). New fact in — not an outcome resolve and not a grounded ChainLock append. Posterior ≠ truth. Use this when you have a new fact to observe before resolve/calibrate. Do not use it for grounded ChainLock append without AKM, resolving an outcome, or ranked recall; use chainlock_append, memory_resolve, or memory_recall instead. Write: additive learn-chain stamp. authorizes_action stays false. Hash-only cards refuse AKM-NO-FACT. Append-only; there is no memory_delete. fact is required (≤160). subject/memory_id/use_case optional. Mutation requires confirm=true (runtime gate) or dry_run=true (preview only, no write). confirm and dry_run stay optional on inputSchema.required. dry_run still returns FG-HALLUC-TOOL, FG-STUB, FG-UNKNOWN-OP, or FG-LOCAL-ONLY when a confirm call would refuse. confirm is consent, not tenant auth, and it does not upgrade shared public-demo isolation. Returns memory_id, observation stamp, and display envelope (belief is not truth).

Input Schema

TableJSON Schema

Name	Required	Description
`fact`	Yes	Fact text clipped to 160 characters. Required on observe. Hash-only cards refuse AKM-NO-FACT.
`confirm`	No	Documented confirmation flag. Optional in inputSchema.required (connector refresh must not break). tools/call still refuses MCP-CONFIRM-REQUIRED when confirm is missing or false unless dry_run=true. confirm is consent to run the call. It is not tenant auth and it does not upgrade shared public-demo isolation.
`dry_run`	No	Optional preview flag. When true, do not write and do not stamp a ledger tip. Allowlist, hallucination, stub, and local-only checks still run and return the same FragGate refuse code a confirm call would get. MCP-DRY-RUN is returned only when that call would have been allowed to proceed. confirm is consent, not tenant auth, and it does not upgrade shared public-demo isolation.
`subject`	No	Optional subject key clipped to 80 characters. Used to find or create memory_id.
`use_case`	No	Optional use-case label for calibration / adaptive recall ranking. Not a truth claim.
`memory_id`	No	Optional existing memory id. Alternative to subject for resolve/calibrate/get.

Output Schema

TableJSON Schema

Name	Required	Description
`code`	No	FragGate or fabric code when present: FG-OK, FG-HALLUC-TOOL, FG-STUB, FG-LOCAL-ONLY, FG-UNKNOWN-OP, FG-GATE-REFUSE, FG-LAMB-REFUSE, or a module refuse such as MESH-* / AKM-*.
`door`	No	Door name. The public door is fraggate.
`ran_in`	No	Execution locale (for example aziel-runtime) when present.
`result`	No	FragGate body: ok, code, door, slug, op, engine_digest, ran_in, provenance, refusal, limitations, receipt, plus the engine result. Unknown names refuse FG-HALLUC-TOOL; stubs refuse FG-STUB.
`status`	No	HTTP-like status when present on wrappers (200 ok; 400+ error / refuse).
`display`	No	Human-facing envelope. Show display.action, title, and summary, then take the next input. Never echo raw tool names.
`receipt`	No	Optional receipt, ledger tip, or TemporalLock/ForgeReceipts exit when the door stamped one.
`refusal`	No	Explicit refuse object, code, or message when the door or engine refused.
`engine_op`	No	Resolved engine op when present (often inside result).
`ledger_tip`	No	Ask/refuse ledger tip when the door stamped one.
`provenance`	No	Provenance / input packet when the pipeline attached one.
`session_id`	No	Raw session id when session plumbing was used. Hidden unless the user asked for the chain.
`engine_slug`	No	Resolved engine slug when present (often inside result).
`limitations`	No	Capability limitations or Remain-OFF notes when present.
`engine_digest`	No	64-hex engine_digest when a true in-process engine ran (often inside result).

Schema Changelog

Changes observed during successful MCP inspections.

Changed5 schema fields changedv2.0.9Sep 29, 2026

changedOutput schema / description

Previous value: -"Display envelope shown to the user (display.title / display.summary) plus the machine result. Extra engine fields may appear."New value: +"Display envelope shown to the user (display.action / display.title / display.summary) plus the machine result. Extra engine fields may appear. Do not echo raw tool names."

changedOutput schema / properties / display / description

Previous value: -"Human-facing envelope. Show title and summary, then take the next input."New value: +"Human-facing envelope. Show display.action, title, and summary, then take the next input. Never echo raw tool names."

addedOutput schema / properties / display / properties / action

Added value: +{
+  "description": "Human-visible run frame. Always \"Run aziel runtime\". Product verb titles stay on title. Not a tool name.",
+  "type": "string"
+}

addedOutput schema / properties / display / properties / image

Added value: +{
+  "additionalProperties": true,
+  "description": "Present only when a real production included png_b64, jpeg_b64 (or the same byte family), or a cited http(s) image URL. No invented thumbs. Omitted on dry_run. reviewed is true only when executed production bytes are attached; a dry_run never mints reviewed.",
+  "properties": {
+    "data": {
+      "description": "Base64 image bytes copied from the production. Omitted for a URL-only cite. Never present on dry_run.",
+      "type": "string"
+    },
+    "mimeType": {
+      "description": "Sniffed image media type (image/png, image/jpeg, image/webp, or image/gif).",
+      "type": "string"
+    },
+    "reviewed": {
+      "description": "True only when data came from an executed production. False for a URL cite without bytes. Never true on dry_run.",
+      "type": "boolean"
+    },
+    "source": {
+      "description": "Production field the image was taken from (for example png_b64, jpeg_b64, or image_url).",
+      "type": "string"
+    },
+    "url": {
+      "description": "Cited http(s) image URL when the production named one. Not invented.",
+      "type": "string"
+    }
+  },
+  "type": "object"
+}

changedOutput schema / properties / display / properties / title / description

Previous value: -"Short result title for the AI client."New value: +"Product verb title for the AI client. Not a raw tool name."

Changed2 schema fields changedv2.0.8Sep 27, 2026

changedInput schema / properties / confirm / description

Previous value: -"Documented confirmation flag. Optional in inputSchema.required (connector refresh must not break). tools/call still refuses MCP-CONFIRM-REQUIRED when confirm is missing or false unless dry_run=true (preview, no write)."New value: +"Documented confirmation flag. Optional in inputSchema.required (connector refresh must not break). tools/call still refuses MCP-CONFIRM-REQUIRED when confirm is missing or false unless dry_run=true. confirm is consent to run the call. It is not tenant auth and it does not upgrade shared public-demo isolation."

changedInput schema / properties / dry_run / description

Previous value: -"Optional preview flag. When true, return a would-mutate preview and do not write. Alternative to confirm=true. Does not mutate."New value: +"Optional preview flag. When true, do not write and do not stamp a ledger tip. Allowlist, hallucination, stub, and local-only checks still run and return the same FragGate refuse code a confirm call would get. MCP-DRY-RUN is returned only when that call would have been allowed to proceed. confirm is consent, not tenant auth, and it does not upgrade shared public-demo isolation."

Changed3 schema fields changedv2.0.2Sep 18, 2026

changedInput schema / description

Previous value: -"fact is required. subject/memory_id/use_case optional."New value: +"fact is required. subject/memory_id/use_case optional. Mutation requires confirm=true or dry_run=true."

addedInput schema / properties / confirm

Added value: +{
+  "description": "Documented confirmation flag. Optional in inputSchema.required (connector refresh must not break). tools/call still refuses MCP-CONFIRM-REQUIRED when confirm is missing or false unless dry_run=true (preview, no write).",
+  "type": "boolean"
+}

addedInput schema / properties / dry_run

Added value: +{
+  "description": "Optional preview flag. When true, return a would-mutate preview and do not write. Alternative to confirm=true. Does not mutate.",
+  "type": "boolean"
+}

Addedv1.6.2Sep 12, 2026

TDQS

A4.4/5.0

Behavior5/5

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

Beyond the annotations, the description discloses append-only semantics ('there is no memory_delete'), the mutation gate (confirm=true or dry_run=true), preview behavior, specific error codes (FG-HALLUC-TOOL, FG-STUB, FG-UNKNOWN-OP, FG-LOCAL-ONLY), and that confirm is consent rather than tenant auth and does not upgrade isolation. There is no contradiction with annotations (readOnlyHint=false is consistent with a mutating tool).

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

Conciseness2/5

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

The description is a dense, single paragraph with heavy domain jargon (AKM-TRIAD-1.0, FG-*, ChainLock, hash-only cards, AKM-NO-FACT). Sentences like 'Posterior ≠ truth' and 'Write: additive learn-chain stamp' are cryptic and do not clearly help an agent decide to call the tool. It repeats information already present in the schema (e.g., fact required, confirmation gate) and lacks clear structure or front-loading, making it harder to parse than necessary.

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

Completeness5/5

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

Given the high complexity of this tool—multiple failure modes, mutation gates, dry-run semantics, and consent caveats—the description is thorough and covers all relevant operational details. It even explains what happens in preview mode, error codes that can be returned, and the limits of confirm as consent. Since an output schema exists, the description does not need to explain return values, and this description is complete enough for an agent to call the tool correctly.

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?

The schema coverage is 100% and the schema itself already provides rich descriptions for every parameter, including the required fact, the confirm/dry_run behavior, and subject/memory_id/use_case. The description mostly repeats this information ('fact is required (≤160). subject/memory_id/use_case optional') and adds little new semantic meaning beyond the schema, so the baseline score of 3 is appropriate.

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

Purpose5/5

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

The description states a specific verb and resource: 'Append the first memory_observation to the ChainLock learn chain'. It explicitly distinguishes this from sibling operations ('not an outcome resolve and not a grounded ChainLock append') and names the tools to use instead (chainlock_append, memory_resolve, memory_recall). An agent can tell this tool apart from its siblings without inspecting schemas.

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

Usage Guidelines5/5

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

The description gives explicit when-to-use guidance ('Use this when you have a new fact to observe before resolve/calibrate') and when-not-to-use guidance with specific alternatives ('Do not use it for grounded ChainLock append without AKM, resolving an outcome, or ranked recall; use chainlock_append, memory_resolve, or memory_recall instead'). This is the gold standard for usage guidance.

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